Download pdf - SuperB Progress Reports - cithep.caltech.edufcp/superB/wp.pdf · SuperB Progress Reports Physics Accelerator Detector Computing March 31, 2010 Abstract This report describes the present

SuperBProgress Reports

Physics

Accelerator

Detector

Computing

March 31, 2010

Abstract

This report describes the present status of the detector design for SuperB. It is one of four separateprogress reports that, taken collectively, describe progress made on the SuperB Project since thepublication of the SuperB Conceptual Design Report in 2007 and the Proceedings of SuperBWorkshop VI in Valencia in 2008.

Contents

1 Introduction 11.1 The Physics Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 The SuperB Project Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3 The Detector Design Progress Report . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2 Overview 22.1 Physics Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.2 Challenges on Detector Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.3 Open Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.4 Detector R&D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

3 Silicon Vertex Tracker 93.1 Detector concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

3.1.1 SVT and Layer0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93.1.2 Performance Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103.1.3 Background Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

3.2 Layer0 options under study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123.2.1 Striplets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123.2.2 Hybrid Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.2.3 MAPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.2.4 Pixel Module Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

3.3 A MAPS-based all-pixel SVT using a deep P-well process . . . . . . . . . . . . . . . 163.4 R&D Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

4 Drift Chamber 194.1 Backgrounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204.2 Mechanical Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204.3 Drift Chamber Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204.4 Gas Mixture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214.5 Cell Design and Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224.6 R&D work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

5 Particle Identification 245.1 Detector concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

5.1.1 Charged particle identification at SuperB . . . . . . . . . . . . . . . . . . . . 245.1.2 BABAR DIRC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

5.2 Barrel PID at SuperB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255.2.1 Performance optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255.2.2 Design and R&D status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

5.3 Forward PID at SuperB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285.3.1 Motivation for a forward PID detector . . . . . . . . . . . . . . . . . . . . . . 285.3.2 Forward PID requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305.3.3 Status of the forward PID R&D effort . . . . . . . . . . . . . . . . . . . . . . 30

6 Electromagnetic Calorimeter 336.1 Barrel Calorimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

6.2 Forward Endcap Calorimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 346.3 Backward Endcap Calorimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366.4 R&D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

6.4.1 Barrel Calorimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396.4.2 Forward Calorimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416.4.3 Backward Calorimeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

7 Instrumented Flux Return 427.1 Performance optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

7.1.1 Identification Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427.1.2 Baseline Design Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . 427.1.3 Design optimization and performance studies . . . . . . . . . . . . . . . . . . 43

7.2 R&D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447.2.1 R&D tests and results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447.2.2 Prototype . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

7.3 Baseline Detector Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467.3.1 Flux Return . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

8 Electronics, Trigger, DAQ and Online 478.1 Overview of the Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

8.1.1 Trigger Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478.1.2 Trigger Rates and Event Size Estimation . . . . . . . . . . . . . . . . . . . . 498.1.3 Dead Time and Buffer Queue Depth Considerations . . . . . . . . . . . . . . 49

8.2 Electronics, Trigger and DAQ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 498.2.1 Fast Control and Timing System . . . . . . . . . . . . . . . . . . . . . . . . . 508.2.2 Clock, Control and Data Links . . . . . . . . . . . . . . . . . . . . . . . . . . 518.2.3 Common Front-End Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . 528.2.4 Readout Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 538.2.5 Experiment Control System . . . . . . . . . . . . . . . . . . . . . . . . . . . . 538.2.6 Level 1 Hardware Trigger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

8.3 Online System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 558.3.1 ROM Readout and Event Building . . . . . . . . . . . . . . . . . . . . . . . . 568.3.2 High Level Trigger Farm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 568.3.3 Data Logging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 568.3.4 Event Data Quality Monitoring and Display . . . . . . . . . . . . . . . . . . . 578.3.5 Run Control System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 578.3.6 Detector Control System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 578.3.7 Other Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 578.3.8 Software Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

8.4 Front-End Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 588.4.1 SVT Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 588.4.2 DCH Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 598.4.3 PID Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 598.4.4 EMC Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 618.4.5 IFR Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

8.5 R&D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 638.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

9 Software and Computing 649.1 The baseline model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

9.1.1 The requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 659.2 Computing tools and services for the Detector and Physics TDR studies . . . . . . . 66

9.2.1 Fast simulation of the SuperB detector . . . . . . . . . . . . . . . . . . . . . . 669.2.2 Bruno: the SuperB full simulation tool . . . . . . . . . . . . . . . . . . . . . . 69

9.3 Simulation output: Hits and MonteCarlo Truth . . . . . . . . . . . . . . . . . . . . . 709.4 Staged simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 709.5 Interplay with FastSim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

9.5.1 The distributed production environment . . . . . . . . . . . . . . . . . . . . . 719.5.2 The software development and collaborative tools . . . . . . . . . . . . . . . . 77

10 Mechanical Integration 79

11 Budget and Schedule 7911.1 Detector Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8011.2 Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

1.2 The SuperB Project Elements 1

1 Introduction

1.1 The Physics Motivation

The Standard Model successfully explains thewide variety of experimental data that hasbeen gathered over several decades with ener-gies ranging from under a GeV up to severalhundred GeV. At the start of the millennium,the flavor sector was perhaps less explored thanthe gauge sector, but the PEP-II and KEK-Basymmetric B Factories, and their associatedexperiments BABAR and Belle, have produceda wealth of important flavor physics highlightsduring the past decade [1]. The most notableexperimental objective, the establishment of theCabibbo-Kobayashi-Maskawa phase as consis-tent with experimentally observed CP-violatingasymmetries in B meson decay, was cited in theaward of the 2008 Nobel Prize to Kobayashi &Maskawa [2].

The B Factories have provided a set of unique,over-constrained tests of the Unitarity Triangle.These have, in the main, been found to be con-sistent with Standard Model predictions. The Bfactories have done far more physics than orig-inally envisioned; BABAR alone has publishedmore than 400 papers in refereed journals todate. Measurements of all three angles of theUnitarity Triangle - sin2α and γ, in addition tosin 2β; the establishment of D0D̄0 mixing; theuncovering of intriguing clues for potential NewPhysics in B→ K(?)l+l− and B→ Kπ and de-cays; and unveiling an entirely unexpected, newspectroscopy, are some examples of importantexperimental results beyond those initially con-templated.

With the LHC now beginning operations, themajor experimental discoveries of the next fewyears will probably be at the energy frontier,where we would hope not only to complete theStandard Model by observing the Higgs parti-cle, but to find signals of New Physics which arewidely expected to lie around the 1 TeV energyscale. If found, the New Physics phenomena willneed data from very sensitive heavy flavor ex-

periments if they are to be understood in detail.Determining the nature of the New Physics in-volved requires the information on rare b, c andτ decays, and on CP violation in b and c quarkdecays that only a very high luminosity asym-metric B Factory can provide [3]. On the otherhand, if such signatures of New Physics are notobserved at the LHC, then the excellent sensi-tivity provided at the luminosity frontier by Su-perB provides another avenue to observing NewPhysics at mass scales up to 10 TeV or morethrough observation of rare processes involvingB and D mesons and studies of LFV in τ decays.

1.2 The SuperB Project Elements

It is generally agreed that the physics being ad-dressed by a next-generation B factory requiresa data sample that is some 50-100 times largerthan the existing combined sample from BABARand Belle, or at least 50-75 ab−1. Acquiring suchan integrated luminosity in a 5 year time framerequires that the collider run at a luminosity ofat least 1036cm−2s−1.

For a number of years, an Italian led, INFNhosted, collaboration of scientists from Canada,Italy, Israel, France, Norway, Spain, Poland, UKand the US have worked together to design andpropose a high luminosity 1036 asymmetric BFactory project, called SuperB to be built at ornear the Frascati laboratory [4]. The project,which is managed by a project board, includesdivisions for the accelerator, the detector, thecomputing, and the site & facilities.

The accelerator portion of the projectemploys lessons learned from modern low-emittance synchrotron light sources andILC/CLIC R&D, and an innovative new ideafor the intersection region of the storage rings[5], called crab waist, to reach luminosities over50 times greater than those obtained by earlierB factories at KEK and SLAC. There is nowan attractive, cost-effective accelerator design,including polarization, which is being furtherrefined and optimized [6]. It is designedto incorporate many PEP-II components.This facility promises to deliver fundamen-

SuperB Detector Progress Report

2 2 Overview

tal discovery-level science at the luminosityfrontier.

There is also an active international proto-collaboration working effectively on the designof the detector. The detector team draws heav-ily on its deep experience with the BABAR de-tector, which has performed in an outstandingmanner both in terms of scientific productivityand operational efficiency. BABAR serves as thefoundation of the upgraded SuperB detector.

To date, the project has been very favorablyreviewed by several international committees.This international community now awaits a de-cision by the Italian government on its supportof the SuperB project.

1.3 The Detector Design ProgressReport

This document describes the design and devel-opment of the SuperB detector, which is basedon a major upgrade of BABAR. This is oneof several descriptive ”Design Progress Reports(DPR)” being produced by the SuperB projectduring the first part of 2010 to motivate andsummarize the development, and present statusof each major division of the project (Physics,Accelerator, Detector, and Computing) so as topresent a snapshot of the entire project at a in-termediate stage between the CDR, which waswritten in 2007, and the TDR that is being de-veloped during the next year.

This ”Detector DPR” begins with a briefoverview of the detector design, the challengesinvolved in detector operations at the luminos-ity frontier, the approach being taken to opti-mize the remaining general design choices, andthe R&D program that is underway to developand validate the system and subsystem designs.Each of the detector subsystems and the generaldetector systems are then described in more de-tail, followed by a description of the integrationand assembly of the full detector. Finally, thepaper concludes with a discussion of detectorcosts and a schedule overview.

References

[1] C. Amsler et al. (Particle Data Group),Physics Letters B667, 1 (2008).

[2] http://nobelprize.org/nobel_prizes/physics/laureates/2008/press.html,and http://www-public.slac.stanford.edu/babar/Nobel2008.htm.

[3] D. Hitlin et al. ”Proceedings of SuperBWorkshop VI: New Physics at the SuperFlavor Factory”; arXiv:0810.1312v2 [hep-ph].

[4] M. Bona et al. ”SuperB : A High-Luminosity Heavy Flavour Fac-tory: Conceptual Design Report”;arXiv:0709.0451v2 [hep-ex], INFN/AE- 07/2, SLAC-R-856, LAL 07-15.

[5] P. Raimondi, 2nd LNF Workshop on Su-perB , Frascati, Italy, March 16-18 2006,and Proceedings of the 2007 Particle Ac-celerator Conference (PAC 2007) Albu-querque, New Mexico, USA, June 25-29.

[6] Design Progress Report for the SuperB Ac-celerator (2010) in preparation.

2 Overview

The SuperB detector concept is based on theBABAR detector, with those modifications re-quired to operate at a luminosity of 1036

or more, and with a reduced center-of massboost. Further improvements needed to copewith higher beam-beam and other beam-relatedbackgrounds, as well as to improve detector her-meticity and performance, are also discussed,as is the necessary R&D required to implementthis upgrade. Cost estimates and the scheduleare described in Section 11.

The current BABAR detector consists of atracking system with a 5 layer double-sided sili-con strip vertex tracker (SVT) and a 40 layerdrift chamber (DCH) inside a 1.5T magnetic


http://nobelprize.org/nobel_prizes/ physics/laureates/2008/press.htmlhttp://nobelprize.org/nobel_prizes/ physics/laureates/2008/press.htmlhttp://www-public.slac. stanford.edu/babar/Nobel2008.htmhttp://www-public.slac. stanford.edu/babar/Nobel2008.htm

2.1 Physics Performance 3

field, a Cherenkov detector with fused silica barradiators (DIRC), an electromagnetic calorime-ter (EMC) consisting of 6580 CsI(Tl) crystalsand an instrumented flux-return (IFR) com-prised of both limited streamer tube (LST) andresistive plate chamber (RPC) detectors for K0Ldetection and µ identification.

The SuperB detector concept reuses a num-ber of components from BABAR: the flux-returnsteel, the superconducting coil, the barrel of theEMC and the fused silica bars of the DIRC.The flux-return will be augmented with addi-tional absorber to increase the number of inter-actions lengths for muons to roughly 7λ. TheDIRC camera will be replaced by multi-channelplate (MCP) photon detectors in focusing con-figuration with fused silica optics to reduce theimpact of beam related backgrounds and im-prove performance. The forward EMC will fea-ture cerium-doped LSO (lutetium orthosilicate)or LYSO (lutetium yttrium orthosilicate) crys-tals, hereafter referred to as L(Y)SO crystals,which have a much shorter scintillation timeconstant, a lower Moliére radius and better ra-diation hardness than the current CsI(Tl) crys-tals, again for reduced sensitivity to beam back-grounds and better position resolution.

The tracking detectors for SuperB will benew. The current SVT cannot operate at L =1036, and the DCH has reached the end of its de-sign lifetime and must be replaced. To maintainsufficient ∆t resolution for time-dependent CPviolation measurements with the SuperB boostof βγ = 0.24, the vertex resolution will be im-proved by reducing the radius of the beam pipe,placing the inner-most layer of the SVT at a ra-dius of roughly 1.2 cm. This innermost layerof the SVT will be constructed of either siliconstriplets or MAPS or other pixelated sensors,depending on the estimated occupancy frombeam-related backgrounds. Likewise the cellsize and geometry of the DCH will be driven byoccupancy considerations. The hermeticity ofthe SuperB detector, and thus its performancefor certain physics channels will be improved byincluding a backwards ”veto-quality” EMC de-tector comprising a lead-scintillator stack. The

justification for inclusion of a forward PID isless clear on balance and remains under study.The baseline design concept is a fast Cherenkovlight- based time-of-flight system.

[WE NEED A NEW FIGURE.]The SuperB detector concept is shown in

Fig. 1. The top portion of this elevation viewshows the minimal set of new detector compo-nents, with the most reuse of current BABARdetector components; the bottom half showsthe configuration of new components requiredto cope with higher beam backgrounds and toachieve greater hermeticity.

2.1 Physics Performance

The SuperB detector design as described in theConceptual Design Report [1] left open a num-ber of questions that have a large impact on theoverall detector geometry. The main ones in-clude estimating the effect of a PID device infront of the forward EMC, the need of an EMCin the backward region, the position of the in-nermost layer of the SVT and its internal geom-etry, the SVT-DCH transition radius, and theamount and distribution of absorber in the IFR.

The study of these options has been per-formed by evaluating the physics reach of a setof benchmark decay channels or the overall per-formance in the reconstruction of charged andneutral particles. To accomplish this task a fastsimulation specifically developed for SuperB hasbeen used (sec. 9), combined with a completeset of analysis tools inherited for the most partfrom the BaBar experiment. The main sourcesof machine background have also been simulatedwith GEANT4 to estimate the rates and occu-pancies as a function of the position. The mainresults of the ongoing performance studies aresummarized in this section.

Time-dependent measurements are an impor-tant part of the SuperB physics program. Tokeep a time resolution comparable to what wasmeasured at BABAR, the SuperB reduced boostmust be compensated with a much better vertexresolution by placing the innermost layer of theSVT (Layer0) as close as possible to the IP. Themain factor limiting the minimum distance from


4 2 Overview

Figure 1: Concept for the SuperB detector. The upper half shows the baseline concept, and thebottom half adds a number of detector optional configurations.

the IP is the hit rate from e+e− → e+e−e+e−background events. In this context the perfor-mances of the hybrid pixels (1.08% X0, 14µmhit reso.) and striplets (0.40% X0, 8µm hitreso.) have been compared. Simulation stud-ies of B0 → ΦK0S decays have shown that witha boost βγ = 0.28 the hybrid pixels and thestriplets reach a sin 2βeff per event error equalto BABAR at a distance of 1.5 cm and 2.0 cm,respectively. With βγ = 0.24 the error in-creases by 7-8%. Similar conclusions also applyto B0 → π+π−. These results will help decid-ing what is the most appropriate technology andposition for the Layer0.

The BABAR SVT five-layer design was moti-vated by the request of standalone tracking forlow-pT tracks and redundancy in case severalmodules failed during operations. The defaultSuperB SVT design consisting of a Layer0 plusa BABAR-like SVT detector has been comparedwith two alternative models made of a total

of 5 or 4 layers. Studies of track parametersresolutions and B → D∗K kinematic variablesand reconstruction efficiency have shown thatwhen the number of layers is reduced the low-pT track efficiency decreases significantly, whilethe tracks quality is basically unaffected. Theseresults support a six-layer layout.

Studies have also shown that the best over-all SVT+DCH tracking performance would beachieved when the outer radius of the SVT iskept small (14 cm as in BABAR or even less) andthe inner wall of the DCH is as close to the SVTas possible. However, though in the SuperB de-tector there is not a fixed support tube as therewas in BABAR, space between SVT and DCHmust be left to allocate the cryostats for thesuper-conducting magnets in the interaction re-gion. This constraint is expected to limit theminimum DCH inner radius to about 20-25 cm.

The impact of a forward PID device has beenestimated analyzing the physics reach in chan-


2.1 Physics Performance 5

nels such as B → K(∗)νν̄ by weighting the ad-vantage of having a better PID information inthe forward region with the drawbacks arisingfrom more material in front of the EM calorime-ter and a slightly shorter DCH. Three detec-tor configurations have been compared: BABAR,the SuperB baseline (no forward PID device),and the configuration with the addition of atime-of-flight (TOF) detector between the DCHand the forward EMC. The results for the de-cay mode B → Kνν̄ with the tag side recon-structed in the semileptonic modes are reportedin Fig. 2. The study shows that moving fromBABAR to the SuperB detector instrumentedwith the TOF device the precision S/

√S +B

increases by about 13%, of which 7-8% arisesfrom the increase of the overall detector accep-tance because of the reduced boost and 5-6%is due to the improved pion/kaon separation inthe forward region. The machine backgroundswere not included in the simulation. The analy-sis will be repeated keeping them into account.

]-1Integrated Lumi[ab10 20 30 40 50 60 70

S/s

qrt

(S+B

)

1

2

3

4

5

6

νν+K→+Gains in Signal B

= 0.56)βγBaBar (

= 0.28)βγSuperB base-line (

base-line+TOF

νν+K→+Gains in Signal B

Figure 2: S/√S +B of B → Kνν̄ as a function

of the integrated luminosity in threedetector configurations.

The backward calorimeter under considera-tion is designed to be used in veto mode. Itsimpact on physics can be estimated by study-ing the sensitivity of rare B decays with oneor more neutrinos in the final state, which ben-efit from having a more hermetic detection ofneutrals to reduce the background contamina-tion. One of the most important benchmark

channels of this kind is B → τν. Preliminarystudies, not including the machine backgrounds,indicate that when the backward calorimeter isinstalled the statistical precision S/

√S +B is

enhanced by about 10%. The results are sum-marized in Fig. 3. The top plot shows howS/√S +B changes as a function of the cut on

Eextra (the total energy of charged and neutralparticles that cannot be directly associated withthe reconstructed daughters of the signal or tagB) with and without the backward EMC. Thesignal is peaked at zero. The bottom plot showsthe ratio of S/

√S +B as a function of the Eextra

cut. The analysis will be repeated including themain sources of machine background, which canaffect the Eextra distributions significantly. Thepossibility of using the backward calorimeter asa PID time-of-flight device is under study.

, GeVextraCut on E0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

-1 @

75

abS

+BS

2

4

6

8

10

12SuperB With BwdNo Bwd

=50000gensigN

=10000000

genbkgN

, GeVextraCut on E0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

w/o

S+BS

/w

ith

S+BS

0.850.9

0.951

1.051.1

1.151.2

1.25SuperB With Bwd/No Bwd

Figure 3: Top: S/√S +B as a function of the

cut on Eextra with (circles) and with-out (squares) the backward EMC.Bottom: ratio of S/

√S +B as a func-

tion of the Eextra cut.

The presence of a forward PID or backwardEMC affects the maximum extension of theDCH and therefore the tracking and the dE/dxperformance in those regions. The impact of theTOF PID detector is practically negligible be-


6 2 Overview

cause it only takes a few centimeters from theDCH. On the other hand, the effect of a forwardRICH device (∼ 20 cm DCH length reduction)or the backward EMC (∼ 30 cm) is somewhatlarger. For example, it is found a σ(p)/p in-crease of about 25% and 35% for tracks withpolar angle of 23◦ and 150◦, respectively. Evenin this case, however, the overall impact is gen-erally quite limited because only a small fractionof tracks cross the extreme forward and back-ward regions.

The IFR system will be upgraded by replac-ing the BABAR’s RPCs and LSTs with layersof much faster extruded plastic scintillator cou-pled to WLS fibers read out by APDs operatedin Geiger mode. The identification of muonsand K0L is optimized with a GEANT4 simula-tion by tuning the amount of iron absorber andthe distribution of the active detector layers.The current baseline design has an iron thick-ness of 92 cm segmented with 8 layers of scin-tillator. Preliminary estimates indicate a muonefficiency larger than 90% for p > 1.5 GeV/cwhen the pion misidentification rate is 2%.

2.2 Challenges on Detector Design

Besides the production of the short lived par-ticles that are the main object of investigationof SuperBmany other phenomena connected tothe collider operation generate long lived parti-cles interacting with the SuperBdetector. Theselatter particles form the machine background.

The problem of the machine backgroundis one of the leading challenges of the Su-perBproject: each subsystem must be designedso that its performances are minimally degradedbecause of the occupancy produced by the back-ground hits moreover the detectors must be pro-tected against deteriorations arising from radi-ation damage.

In effect, what is required is to achieve detec-tor performances and operational lifetimes sim-ilar or better than those achieved in BABAR butat a two order of magnitude higher luminosity.

Backgrounds particles produced by beam gasscattering and by the synchrotron radiationnear the interaction point (IP) are expected to

be manageable since the relevant SuperBdesignparameter (mainly the beam current) is fairlyclose to the PEP-II one.

Touschek backgrounds are expected to belarger than in BABAR since the extremely lowdesign emittances of the SuperBbeams. Pre-liminary simulations indicate that a system ofbeam collimators upstream the IP can reducethe particles losses at tolerable levels.

The main source of concern arise from thebackground particles produced at the IP byQED processes whose cross section is order of200 mb that corresponds at nominal luminosityto a rate order of 200 GHz. The main process isthe radiative Bhabha one (i.e.: e+e− → e+e−γ)in which one of the incoming beam particleslooses a significant fraction of its energy by theemission of a photon. Both the photon andthe radiating beam particles emerge from theIP traveling almost collinearly with respect tothe beam-line. The magnetic elements down-stream the IP over-steers these primaries parti-cles into the vacuum chamber walls producingelectro-magnetic showers whose final productsare the backgrounds particles seen by the sub-system. The particles of these electromagneticshowers can also excite the giant nuclear res-onance of the material around the beam lineexpelling neutrons from the nucleus. A carefuloptimization of the mechanical aperture of thevacuum chambers and of the optical functionsis needed to keep a large stay clear for the off-energy primaries particles hence reducing thebackground rate.

The first Geant4 full Monte Carlo simula-tions of this process at SuperB indicates that ashield around the beam line will be required tokeep the electrons, positrons, photons and neu-trons away from the detector both to keep occu-pancies and radiation damage at a comfortablelevel.

Besides the “radiative Bhabha” the “quasielastic Bhabha” process was also considered.The cross section for producing a primary par-ticle reconstructed by the detector via this pro-cess is order of 100 nb that correspond to a rateorder of 100 kHz. It is reasonable to assume


2.3 Open Issues 7

that this will be the driving term of the levelone trigger rate. Single beam contributions tothe trigger rate are in fact expected to be of thesame order of the BABAR one being the nomi-nal beam currents and the other relevant designparameters comparables.

The final issue related to high luminosityis the production of electron positron pairs atthe IP by the two photons process e+e− →e+e−e+e− whose total cross section evaluatedat leading order with the Monte Carlo genera-tor DIAG36 [2] is 7.3 mb that corresponds atnominal luminosity to a rate of 7.3 GHz. Thepairs produced by this process are characterizedby a very soft transverse momentum spectrum.The solenoidal magnetic field in the trackingvolume confines most of these background par-ticles inside the beam pipe. The particles hav-ing a transverse momentum greather enough toreach the beam pipe ( pt > 2.5 MeV/c) and witha momentum polar angle inside the layer 0 ac-ceptance are produced with a rate ∼ 0.5 GHz atnominal luminosity. This background will drivethe segmentation size and the read-out architec-ture of the SVT layer 0. The background traksurface rate on the SVT layer 0 as a function ofits radius is reported on Fig.4. An effort to sim-

< 1.3)Helix diameter (cm) @ 1.5 T (-1.3 < 0.5 1 1.5 2 2.5 3

2 s

/cm

µCu

mul

ativ

e pa

rticl

es /

1

-110

1

10

210

Figure 4: Pairs background track rate for unitsurface as a function of the SVTlayer 0 radius. Multiple track hits nottaken into account.

ulate all these backgrounds with a Geant4 basedcode is underway at present. A fairly accurate

model of the detector model and of the beamline elements is available to the collaboration.Several configurations have been simulated andstudied providing some preliminaries guidelinesto the detector and machine teams. The final-ization of the interaction region and detectordesign will require further developments of theGeant4 background simulation tools on the de-tector response side.

2.3 Open Issues

The basic geometry, structure and physics per-formance of the SuperB detector is predeter-mined, in the main, by the retention of theoverall magnet, return steel, and support struc-ture from the BaBar detector, and a number ofits largest, and most expensive, subsystems. Infact, even though this fixes both the basic ge-ometry, and much of the physics performance,it does not really constrain the expected per-formance of the SuperB detector in any impor-tant respect. BaBar was already a fully opti-mized B-factory detector for physics, and anyimprovements in performance that could comefrom changing the overall layout or rebuildingthe large subsystems would be modest over-all. The primary challenge for SuperB is toretain physics performance similar to BaBar inthe higher background environment described insubsection 2.2, while operating at much higher( x50) data taking rates.

Within this overall constraint, optimizationof the geometrical layout and new detector el-ements for the most important physics chan-nels remains of substantial interest. The pri-mary tools for sorting through the options are(1) simulation, performed under the auspices ofa ”Detector Geometry Working Group”, thatstudies basic tracking, PID, and neutrals per-formance of different detector configurations, in-cluding their impact on each other, and studiesthe physics reach for a number of benchmarkchannels; and (2) detector R&D, including pro-totyping, developing new subsystem technolo-gies, and understanding the costs, and robust-ness of systems, as well as their impacts on eachother. The first item, discussed in subsection


8 2 Overview

2.1, clearly provides guidance to the second, asdiscussed in subsection 2.4 and the subsystemchapters which follow, and vice versa.

At the level of the overall detector, the imme-diate issue is to define the detector envelopes.Optimization can and will continue for sometime yet within each detector system. The stud-ies performed to date leave us with the defaultdetector proposal, with only a few open optionsremaining at the level of the detector geometryenvelopes. These open issues are: (1) whetherthere is a forward PID detector, and, if so, atwhat z location does the DCH end and the EMCbegin, and (2) whether there is, or is not, a back-ward EMC. These open issues are expected tobe resolved by the Technical Board within thenext few months following further studies by theDetector Geometry Working Group, in collabo-ration with the relevant system groups.

2.4 Detector R&D

The SuperB detector concept rests, for the mostpart, on well validated basic detector technol-ogy. Nonetheless, each of the detectors has maychallenges due to the high rates and demandingperformance requirements with R&D initiativesongoing in all detector systems to improve thespecific performance, and optimize the overalldetector design. These are described in moredetail in each subsystem section.

The SVT innermost layer has to provide goodspace resolution while coping with high back-ground. Although silicon striplets are a viableoption at moderate background levels, a pixelsystem would certainly be more robust againstbackground. Keeping the material in a pixelsystem low enough not to deteriorate the vertex-ing performance is challenging, and there is con-siderable activity to develop thin hybrid pixelsor, even better, monolithic active pixels. Thesedevices may be part of planned upgrade pathand installed as a second generation layer 0. Ef-forts are directed towards the development ofsensors, high rate readout electronics, coolingsystems and mechanical support with low ma-terial content.

In the DCH, many parameters must be op-timized for SuperB running, such as the gasmixture and the cell layout. Precision measure-ments of fundamental gas parameters are ongo-ing, as well as studies with small cell chamberprototypes and simulation of the properties ofdifferent gas mixtures and cell layouts. A possi-ble improvement of the performance of the DCHis the innovative Cluster Counting method, inwhich single cluster of charge are resolved time-wise and counted, improving the resolution onthe track specific ionization and the space accu-racy. This technique requires significant R&Dto be proven feasible in the experiment.

Though the Barrel PID system takes over ma-jor components from BaBar, the camera andreadout are a significant step forward requiringextensive R&D. The challenges include the per-formance of pixelated PMTs for DIRC, the de-sign of the fused silica optical system, the cou-pling of the fused silica optics to the existingbar boxes, the mechanical design of the cam-era, and the choice of electronics. Many of theindividual components of the new camera arenow under active study by members of the PIDgroup , and runs are underway with a single barprototype located in a cosmic ray telescope. Afull scale (1/12 azimuth ) prototype incorporat-ing the complete optical design is planned forcosmic ray tests during the next two years.

End cap PID devices are less well understood,and whether or not they are well motivatedfor the overall detector remains to be demon-strated. Present R&D is centered on develop-ing a good conceptual understanding of differ-ent proposed concepts, on simulating how theirperformance effects the physics performance ofthe detector, and on conceptual R&D for com-ponents of specific devices to validate conceptsand highlight the technical and cost issues.

The EMC barrel is a well understood deviceat the lower luminosity of BaBar. Though therewill be some technical issues associated with re-furbishing, the main R&D needed at present isto understand the effects of pile-up in simula-tion, so as to be able to design the appropriatefront-end shaping time for the readout.


9

The forward and backward EMCs are bothnew devices, using cutting edge technology.Both will require one or more full beam tests,hopefully at the same time, within the next yearor two. Prototypes for these test are being de-signed and constructed.

Systematic studies of IFR system componentshave been performed in a variety of bench andcosmic ray tests, leading to the present proposeddesign. This design will be beam tested in a fullscale prototype currently being prepared for aFermilab beam. This device will demonstratethe muon identification capabilities as a func-tion of different iron configurations, and willalso be able to study detector efficiency and spa-tial resolution.

At present, the Electronics, DAQ, and Trig-ger (ETD), have been designed for the baseluminosity of 1x1036cm−2sec−1, with adequateheadroom. Further R&D must continue in orderto understand the requirements at a luminosityup to 4 times greater, and to insure that there isa smooth upgrade path when the present designis inadequate. On a broad scale, as discussedin the system chapter, each of the many com-ponents of ETD have numerous technical chal-lenges that will require substantial R&D as thedesign advances.

References

[1] SuperB Conceptual Design Report,http://www.pi.infn.it/SuperB/CDRarXiv:0709.0451v2 [hep-ex]

[2] F. A. Berends, P. H. Daverveldt andR. Kleiss, “Monte Carlo Simulation of TwoPhoton Processes. 2. Complete Lowest Or-der Calculations for Four Lepton Produc-tion Processes in electron Positron Colli-sions,” Comput. Phys. Commun. 40, 285(1986).

3 Silicon Vertex Tracker

3.1 Detector concept

3.1.1 SVT and Layer0

The main task of the Silicon Vertex Tracker is toprovide precise position information on chargedtracks to perform measurement of time- depen-dent CP asymmetries in B0 decays, which formthe basis of the SuperB scientic program, as itdid for the frst generation of asymmetric B Fac-tories. In addition, charged particles with trans-verse momenta lower than 100 MeV/c will notreach the central tracking chamber, so for theseparticles the SVT must provide the completetracking information.

These goals have been reached in the BABARdetector with a five layer of silicon strip de-tectors, shown schematically in Fig. 5. TheBaBar SVT showed excellent performance forthe whole life of the experiment, thanks to a ro-bust design that took into account the physicsrequirements as well as enough safety margin,to coope with the machine background, and re-dundancy considerations. The SuperB SVT de-sign is based on the BaBar vertex detector lay-out with the addition of a an innermost layervery close to the IP (Layer0). This new layer isneeded, with the reduced beam energy asymme-try, to improve the vertex resolution and to keepa time resolution for CP measurement compa-rable to what was measured at BaBar. Physicsstudies and background conditions, as explainedin detail in the next two sections, set stringentrequirements on the Layer0 design: radius ofabout 1.5 cm, high granularity (50 × 50µm2pitch), low material budget (about 1% X0), ad-equate radiation resistance.

For the Technical Design Report preparationseveral options are under study for the Layer0technology, with different levels of maturity, ex-pected performance and safety margin againstbackground conditions: striplets modules basedon high resistivity sensors with short strips, hy-brid pixels and other thin pixel sensors based onCMOS Monolithic Active Pixel Sensor (MAPS).


http://www.pi.infn.it/SuperB/CDR

10 3 Silicon Vertex Tracker

580 mm

350 mrad520 mrad

ee +-

Beam Pipe

Space Frame

Fwd. support cone

Bkwd.supportcone

Front end electronics

Babar

SuperB Beam Pipe

SuperB Layer0

Figure 5: Longitudinal section of the SVT

The current baseline confguration for theLayer0 is based on the striplets option, bee-ing the one that gives the better physics perfor-mance, as detailed in next section. Neverthelessoptions with pixel sensors, more robust in highbackground conditions, are beeing developed,with specifc R&D programs, in order to meetall the Layer0 requirements (i.e. low pitch andmaterial budget, high readout speed and radia-tion hardness). This will allow the replacementof the Layer0 striplets modules in a ”secondphase” of the experiment. For this purpousethe SuperB interaction region and the SVT me-chanics will be designed to ensure a rapid accessto the detector for a fast replacement procedureof the Layer0.

The external SVT layers (1-5), with a ra-dius between 3 and 15 cm, will be build withthe same technology used for the BaBar SVT(double sided silicon strip sensor), that is ade-quate with the machine background conditionsexpected at that location in the SuperB accel-erator scheme (i.e. with low beam currents).

The SVT angular acceptance, constrained bythe interaction region design, will be 300 mradin both the forward and backward directions,corresponding to a solid angle coverage of 95%in the center-of-mass frame.

3.1.2 Performance Studies

The SuperB interaction region design is charac-terized by the small size of the transversal sec-tion of the beams, at the level of few µm forσx, and hundreds of nm for σy. Therefore itwill be possible to reduce the radial dimensionof the beampipe tube, to 1 cm, while prevent-ing the beams to scatter into the beampipe ma-terial within the detector coverage angle. Thetotal amount of radial material of the berilliumbeampipe, which includes a few µm of gold foil,and a water cooling channel, has been estimatedto be less than 0.5% X0. For the proposedvalue for the center of mass boost in SuperB,βγ = 0.28 (7 GeV e− beam against a 4 GeV e+

beam), the average B vertex separation alongthe z coordinate, 〈∆z〉 ' βγcτB = 125µm, isreduced almost by a half with respect to theBABAR experiment, where βγ = 0.55. In or-der to maintain a suitable resolution on ∆t fortime-dependent analyses, the proper time differ-ence between the two B decays, it is necessaryto improve the vertex resolution (about a factor2 better) with respect to the current BABARperformances: typically 50− 80 µm in z for ex-clusively reconstructed modes and 100−150 µmfor inclusively reconstructed modes. The vertexprecision requirements for physics, have beenachieved in the BABAR experiment, thanks tothe performances of the silicon vertex tracker(SVT), a five-layer double-sided silicon detec-


3.1 Detector concept 11

tor. The configuration of the SuperB interac-tion region allows to measure the first hit of thetracks near the production vertex, by adding avertex detector layer (Layer0) very close to thebeampipe and keeping the BABAR SVT layoutfor the outer layers. This six-layer vertex de-tector solution would improve significantly thetrack parameter determination, matching themore demanding requirements on the vertexresolution, while maintaining the stand-alonetracking capabilities for low momentum parti-cles.

The choice among the various options underconsideration for the Layer0 (striplets, CMOSMAPS and hybrid pixels) has to take into ac-count the physics requirements for the vertexresolution, depending on the pitch and the to-tal amount of material of the modules. In ad-dition, to assure optimal performance for trackreconstruction, the sensor occupancy has to bemaintained under the level of a few percent, im-posing further requirements on the sensor seg-mentation and on the front-end electronics. Ra-diation hardness should also be taken into ac-count, although it is expected not to be par-ticularly demanding compared to LHC detectorspecifications.

In order to simulate the resolution on the Bdecay vertices and on ∆t for different Layer0configurations, we have used the fast simula-tion program FastSim [?], which reproduces thedetector response according to analytical pa-rameterizations. Several studies have been per-formed where we have reconstructed exclusivelyone B of the event (Breco) and evaluated theother B (Btag) decay vertex using the chargedtracks of the rest of the event after rejectinglong-lived particles and tracks not compatiblewith the candidate vertex. We have consid-ered different B decay modes as Breco, suchas B → π+π−, φK0S and also decay modeswhere the impact of the Layer0 on the decayvertex determination is less effective, such asB → K0SK0S , K0Sπ0. For each decay mode wehave studied the resolution on ∆t and the per-event error on the physical interesting quantitysin(2βeff ).

Figure 6: Resolution on the proper time differ-ence of the two B mesons (βγ = 0.28),for different Layer0 radii, as a functionof Layer0 thickness (in X0 %).

The main result is that the resolution on ∆tat SuperB allows a comparable or even betterper-event error on sin(2βeff ) for each B decaymode that we have considered. The conclusionis valid for all the technologies that we have con-sidered for Layer0, (i.e., MAPS, striplets, Hy-brid Pixels) and for reasonable values of theLayer0 radius and amount of radial material.As an example, in Fig. 6 is reported the resolu-tion on ∆t for different Layer0 radii as a func-tion of the Layer0 thickness (in X0 %) comparedto the BaBar reference value. The dashed linerepresents the BABAR reference value using thenominal value of the boost, βγ = 0.55, accord-ing to FastSim.

We have also studied the impact of a possibleLayer0 inefficiency on the sensitivity on sin(2β).The source of inefficiency could be related toseveral causes, for example a much higher back-ground rate than expected causing dead time inthe readout of the detector. In Fig. 7 is reportedthe sin(2βeff ) per-event error for the B → φK0Sdecay mode as a function of the Layer0 hit effi-ciency for the different options (i.e. differentmaterial budget). The Layer0 radius in thestudy is about 1.6 cm. As it is evident fromthe plot, the striplet solution allows for better



Hybrid pixels

Striplets

MAPS

Figure 7: sin2βeff per event error as a funcionof the Layer0 effciency for the dif-ferent options (i.e. different materialbudget).

performances with respect to the BaBar refer-ence value even in case of small inefficiency, andit has better performances compared to otherLayer0 solutions. The main advantage of thestriplet solution is the smaller amount of radialmaterial (about 0.5 % X0) compared to the Hy-brid pixel (about 1% X0) and the MAPS solu-tions (about 0.7 % X0). Infact, for particles ofmomentum up to few GeV/c the multiple scat-tering effect is the dominant source of uncer-tainty in the determination of their trajectoryand a low material budget detector reduces thiseffect. A striplet-based Layer0 solution wouldhave also a better intrinsic hit resolution (about8 µm) with respect to the MAPS and the Hy-brid Pixel (about 14 µm with a digital readout)solutions. For those reasons a Layer0 based onstriplets has been chosen as the baseline solutionfor SuperB, capable to cope with the machinebackground according to the present estimates.

3.1.3 Background Conditions

Background considerations influence several as-pects of the Silicon Vertex Tracker design:readout segmentation, electronics shaping time,data transmission rate and radiation hardness.Severe requirements are expecially imposed on

the Layer0 design. The different sources ofbackground have been simulated with a detailedGeant4 detector and beamline description to es-timate their impact on the experiment [1]. Thebackground expected in the external layers ofthe SVT (radius > 3 cm) is dominated by termsthat scale with beam currents and is similarto background seen in the present BaBar SVT.The main background at the Layer0 radius isdominated by luminosity terms, in particularby e+e− → e+e−e+e− pair production, beingradiative Bhabha events an order of magnitudesmaller. Despite the huge cross section of thepair production process, the rate of tracks hit-ting Layer0 is strongly suppressed by the effectof the 1.5 Tesla magnetic field inside the de-tector. Particles produced, with low transversemomentum, loop in the detector magnetic fieldand only a small fraction reaches the SVT lay-ers, with a strong radial dependence.

According to these studies the track rate atthe Layer0 at 1.5 cm is at the level of about 5MHz/cm2, mainly due to electrons in the MeVenergy range. The equivalent fluence corre-sponds toabout 3.5x1012n/cm2/yr and the doserate to whitstand is 3 Mrad/yr. It seems ade-quate working with a safety factor of five on thisbackground estimate.

3.2 Layer0 options under study

The present status for the development of thevarious Layer0 options under study for the Tech-nical Design Report preparation is described inthis section.

3.2.1 Striplets

Double-sided silicon strips detectors (DSSD),200µm thick, with 50µm readout pitch repre-sent a proven and robust technology meetingthe requirements on the SVT Layer 0 design, asdescribed in the CDR [1]. In this design, shortstrips will be placed at an angle of ±45◦ to thedetector edge on the two sides of the sensor, asshown in Fig 8.

The strips will be connected to the readoutelectronics through a a multilayer flexible cir-


3.2 Layer0 options under study 13

Figure 8: Schematic view of the two sides of thestriplets detector.

cuit glued to the sensor. A standard technologywith copper traces is already available, althoughan aluminum microcable technology is being ex-plored to reduce the impact on material of theinterconnections.

Figure 9: Mechanical structure of a stripletsLayer 0 module.

The data-driven, high-bandwidth FSSR2readout chip [3], is a good match to the Layer 0striplet design and is also suitable for the read-out of the outer layers strip sensors. It has128 analog channels providing a sparsified dig-ital output with address, timestamp and pulse

height information for all hits. The selectableshaper peaking time can be programmed downto 65 ns. The chip has been realized in a 0.25µmCMOS technology for high radiation tolerance.The readout architecture has been designed tooperate with a 132 ns clock that will define thetimestamp granularity and the readout window.A faster readout clock (70 MHz) is used in thechip, with a token pass logic, to scan for thepresence of hits in the digital section, and totransmit them off-chip, using a selectable num-ber of output data lines. With six output lines,the chip can achieve an output data rate of 840Mb/s. With a 1.83 cm strip length the expectedoccupancy in the 132 ns time window is about12%, considering a hit rate of 100MHz/cm2,that includes the cluster multiplicity and a fac-tor 5 safety margin on the simulated backgroundtrack rate. The FSSR2 readout efficiency hasnever been measured with this occupancy. Firstresults from ongoing Verilog simulations indi-cate the efficiency is 90% or less. As shown inFig. 7 the physics impact such an efficiency ismodest. Nonetheless it may be possible to re-design the digital readout of the FSSR2 to in-crease the readout efficiency at high occupancy.A total equivalent noise charge of 600 e− rms isexpected, including the effects of the strip andflex circuit capacitance, as well as the metal se-ries resistance. The signal to noise for a 200µmdetector is about 26, providing a good noisemargin. It is also foreseen to conduct a mar-ket survey to evaluate whether different readoutchips, possibly with a triggered readout archi-tecture, may provide better performance.

Because of the unfavorable aspect ratio of thesensors, the readout electronics needs to be ro-tated and placed along the beam axis, outside ofthe sensitive volume of the detector, held by acarbon fiber mechanical structure, as shown inFig.9. The 8 modules forming Layer 0 will bemounted on flanges containing the cooling cir-cuits. For the baseline design with striplets, theLayer 0 material budget will be about 0.46%X0for perpendicular tracks, assuming a silicon sen-sor thickness of 200µm, a light module supportstructure ( 100µm Silicon equivalent), similar



to the one used for the BABAR SVT modules,and the multilayer flex contribution (3 flex lay-ers/module, 45µm Silicon equivalent/layer).A reduction in the material budget to about0.35%X0 is possible if kapton/aluminum micro-cable technology can be employed with a tracepitch of about 50µm.

3.2.2 Hybrid Pixels

Hybrid pixels are a mature and viable solutionbut still requires some R&D to meet Layer0 re-quirements (reduction in the front-end pitch andin the total material budget, with respect to hy-brid pixel systems developed for LHC experi-ments) A front-end chip for hybrid pixel sensorwith 50 × 50 µm2 pitch and a fast readout isunder development. The adopted readout ar-chitecture has been previously developed by theSLIM5 Collaboration [4] for CMOS Deep NWellMAPS [5],[6]: the data-push architecture fea-tures data sparsification on pixel and timestampinformation for the hits. This readout has beenrecently optimized for the target Layer0 rate of100 MHz/cm2 with promising results: VHDLsimulation of a full size matrix (1.3 cm2) giveshit efficiency above 98% operating the matrixwith a 60 MHz readout clock. A first proto-type chip with 4k pixels has been submitted inSeptember 2009 with the ST Microelectronics130 nm process and is currently under test. Thefront-end chip, connected by bump-bonding toan high resistivity pixel sensor matrix, will bethen characterized with beams in Autumn 2010.

3.2.3 MAPS

CMOS MAPS are a newer and more challengingtechnology. Their main advantage with respectto hybrid pixels is that they could be very thin,having the sensor and the readout incorporatedin a single CMOS layer, only a few tenth of mi-crons thick. As the readout speed is anotherrelevant aspect for application in the SuperBLayer0 we proposed a new design approach toCMOS MAPS [5] which for the first time madeit possible to build a thin pixel matrix featur-ing a sparsified readout with timestamp infor-

Figure 10: The DNW MAPS concept.

mation for the hits [6]. In this new design thedeep N-well (DNW) of a triple well commercialCMOS process is used as charge collecting elec-trode and is extended to cover a large fractionof the elementary cell (Fig. 3.2.3). Use of a largearea collecting electrode allows the designer toinclude also PMOS transistors in the front-end,therefore taking full adavantage of the proper-ties of a complementary MOS technology forthe design of high performance analog and dig-ital bocks. However, in order to avoid a signif-icant degradation in charge collection effciency,the area covered by PMOS devices and their N-wells, acting as parasitic collection centers, hasto be small with respect to the DNW sensorarea. Note that use of a charge preamplifier asthe input stage of the channel makes the chargesensitivity independent of the detector capaci-tance. The full signal processing chain imple-mented at the pixel level (charge preamplifier,shaper, discriminator and latch) is partly real-ized in the p-well physically overlapped with thearea of the sensitive element, allowing the de-velopment a complex in-pixel logic with similarfunctionalities as in hybrid pixels.

Several prototype chips (the “APSEL” series)have been realized with the STMicroelectron-ics, 130 nm triple well technology and provedthe proposed approach is very promising forthe realization of a thin pixel detector. TheAPSEL4D chip, a 4k pixel matrix with 50× 50µm2 pitch, with a new DNW cell and the spar-sified readout has been characterized duringthe SLIM5 testbeam showing encouraging re-


3.2 Layer0 options under study 15

sults [7]. Hit efficiency of 92% has been mea-sured, a value compatible with the present sen-sor layout that is designed with a fill factor (i.e.the ratio of the electrode over the total n-wellarea) of about 90%. Margins to improve the de-tection efficiency with a different sensor layoutare beeing currently investigated [8]

Several issues still need to be solved todemonstrate the ability to build a working de-tector with this technology and require someR&D. Among others the scalability to largermatrix size and the radiation hardness of thetechnology are under evaluation for the TDRpreparation.

3.2.4 Pixel Module Integration

To minimize the detrimental effect of multiplescattering the reduction of the material is cru-cial for all the components of the pixel modulein the active area.

The pixel module support structure needs toinclude a cooling system to evacuate the powerdissipated by the front-end electronics, about2W/cm2, present in the active area. The pro-posed module support will be realized with alight carbon fiber support with integrated mi-crochannels for the coolant fluid (total materialbudget for support and cooling below 0.3 % X0).Measurements on first support prototypes real-ized with this cooling technique indicate that acooling system based on microchannels can bea viable solution to the thermal and structuralproblem of Layer0 [10].

The pixel module will also need a light multi-layer bus (Al/kapton based with total materialbudget of about 0.2 % X0), with power/signalinputs and high trace density for high dataspeed to connect the front-end chips in the ac-tive area to the HDI hybrid, in the periphery ofthe module. With the data push architecturepresently under study and the high backgroundrate expected data with a 160 MHz clock needto be transfered on this bus. With triggeredreadout architecture (beeing also investigated)the complexity of the pixel bus, and materialassociated, will be reduced.

Figure 11: Schematic drawing of the full Layer0made of 8 pixel modules mountedaround the beam pipe with a pin-wheel arrangement.

Considering the various pixel module compo-nents (sensor and front-end with 0.4% X0, sup-port with cooling, and multilayer bus with de-coupling capacitors) the total material in the ac-tive area for the Layer0 module design based onhybrid pixel is about 1% X0. For a pixel moduledesign based on CMOS MAPS, where the con-tribution of the sensor and the integrated read-out electronics become almost negligible, 0.05%X0, the total material budget is about 0.65% X0.A schematic drawing of the full Layer0 madeof 8 pixel modules mounted around the beampipe with a pinwheel arrangement is shown inFig. 3.2.4.

Due to the high background rate at theLayer0 location radiation-hard fast links be-tween the pixel module and the DAQ systemlocated outside the detector should be adopted.For all Layer0 options (that currently sharea similar data push architecture) the untrig-gered data rate is 16 Gbit/s/readout section, as-suming a background hit rate of 100Mhz/cm2.Triggered data rate is reduced to about 1Gbit/s/readout section.

The HDI. positioned at the end of the mod-ule, outside the active area, will be designed tohost several IC components: some glue logic,buffers, fast serializers, drivers. The compo-



nents should be radiation hard for the applica-tion at the Layer0 location (several Mrad/yr).

The baseline option for the link between theLayer0 modules and the DAQ boards is cur-rently based on a mixed solution. A fast cop-per link is forseen between the HDI and an in-termediate transition board, positioned in anarea of moderate radiation levels (several tens ofkrad/yr). On this transition card the logic withLV1 buffers will store the data until the recep-tion of the LV1 trigger signal and only triggereddata will be send to the DAQ boards with an op-tical link of 1 Gibt/s. The various pixel moduleinterfaces will be characterized in a test set-upfor the TDR preparation.

3.3 A MAPS-based all-pixel SVT using adeep P-well process

Another alternative under evaluation is to havea all-pixel SVT using MAPS pixels with a pixelsize of 50x50 µm. This approach uses the180 nm INMAPS process which incorporates adeep P-well. A perceived limitation of standardMAPS is not having full CMOS capability as theadditional N-wells from the PMOS transistorsparasitically collect charge, thus reducing thecharge collected by the readout diode. Avoid-ing the usage of PMOS transistors however doeslimit the capability of the readout circuitry sig-nificantly. A limited use of PMOS is allowedwith the DNW MAPS design (APSEL chips),which anyway accounts for a small degradationin the collection efficiency. Therefore, a specialdeep P-well layer was developed to overcomethe problems mentioned above. The deep P-well protects charge generated in the epitaxiallayer from being collected by parasitic N-wellsfor the PMOS. This then ensures that all chargeis being collected by the readout diode and max-imises charge collection efficiency. This is illus-trated in Figure 12. This enhancement allowsthe use of full CMOS circuitry in a MAPS andopens completely new possibilities for in-pixelprocessing. The TPAC chip [11] for CALICE-UK [12, 13] has been designed using the IN-MAPS process. The basic TPAC pixel has asize of 50 × 50 µm and comprises a preampli-

fier, a shaper and a comparator [11]. The pixelonly stores hit information in a Hit Flag. Thepixel is running without a clock and the timinginformation is provided by the logic queryingthe Hit Flag. For the SuperB application thepixel design was slightly modified. Instead ofjust a comparator, a peak-hold latch was addedto store the analog information as well. Thechip is organised in columns with a commonADC at the end of each column. The ADC isrealised as a Wilkinson ADC using a 5 MHzclock rate. The simulated power consumptionfor each individual pixel is 12 µW. The columnlogic constantly queries the pixels, but only digi-tises the information for the pixels with a ”HitFlag”. This allows one to save both space andreduce the power usage and since the speed ofthe chip is limited by the ADC also increases thereadout speed. Both the address of the pixel be-ing hit and its ADC output are stored in a FIFOat the end of the column. To further increasethe readout speed, the ADC uses a pipelined ar-chitecture with 4 analog input lines to increasethroughput of the ADC. One of the main bot-tlenecks is getting the data off the chip. It isenvisaged to use the Level 1 trigger informa-tion to reject most of the events and to reducethe data rate on-chip before moving it off-chip.This will significantly reduce the data rate andtherefore also the amount of power and servicesrequired .

For the outer layers, the requirements aremuch more relaxed in terms of occupancy, so inorder to reduce the power, it is planed to mul-tiplex the ADC’s to let them handle more thanone column in the sensor. This is possible dueto the much smaller hit rate in the outer layersand the resulting relaxed timing requirements.

An advantage of the MAPS is the eliminationof a lot of readout electronics, because every-thing is integrated in the sensor already whuchsimplifies the assembly significantly. Also sincewe are using a industry CMOS processs, therea significant price advantage compared to stan-dard HEP-style silicon and the additinal savingsdue to the elimination of a dedicated readoutASIC.


3.4 R&D Activities 17

(a) CMOS MAPS without a deep P-well implant (b) CMOS MAPS with a deep P-well implant

Figure 12: A CMOS MAPS without a deep P-well implant (left) and with a deep P-well implant(right).

In order to evaluate the physics potential ofMAPS based all-pixel vertex detector we arecurrently evaluating the performance of the Su-perB detector with different geometries of theSVT , ranging from the SuperB baseline (Layer0+ 5 layers based on strip detectors), through toa 4 or 6 layer all-pixel detector with a realisticmaterial budget for the support structure for alllayers.

3.4 R&D Activities

The technology for the Layer 0 baseline stripletdesign is well-established but the front-end chipto be used, due to the high background occu-pancy expected, requires some deeper investi-gation. Performance of the FSSR2 chip, pro-posed for the readout of the striplets and theouter layer strip sensors, are beeing evaluated asa function of the occupancy with Verilog simu-tion. Measurements are also possible in a test-bench in preparation with real striplets modulesreadout with the FSSR2 chips. The redesign ofthe digital readout of the chip will be investigateto improve its efficiency. The modification ofthe analog part of the chip for the readout of thelong module of the external layers are currentlyunder study. The multilayer flexible circuit, toconnect the striplets sensor to the frontend, may

benefit from some R&D to reduce the materialbudget: either reduce the minimum pitch onthe Upilex circuit, or adopt kapton/aluminummicrocables and Tape Automated Bonding sol-dering techniques with a 50µm pitch.

Although silicon striplets are a viable optionat moderate background levels, a pixel systemwould certainly be more robust against back-ground. Keeping the material in a pixel systemlow enough not to deteriorate the vertexing per-formance is challenging, and there is consider-able activity to develop thin hybrid pixels or,even better, monolithic active pixels. These de-vices may be part of planned upgrade path andinstalled as a second generation layer 0.

A key issue for the readout of the pixel in theLayer0 is the development of a fast readout ar-chitecture to cope with a pixel rate of the orderof 100MHz/cm2. A first front-end chip for hy-brid pixel sensor with 50 × 50 µm2 pitch anda fast readout, data driven with timestamp forthe hits, has been realized and is currently un-der test. A further development of the architec-ture is beiing pursued to evolve toward a trig-gered readout architecture, hellpful to reducethe complexity of the pixel module and possi-bly to reduce its material budget.


18 References

The CMOS MAPS technology is very promis-ing for an alternative design of the Layer 0,but extensive R&D is still needed to meet allthe requirements. Key aspects to be addressedare: sensor efficiency and its radiation toler-ance, power consumption, and as in the hybridpixel, the readout speed of the architecture im-plemented.

After the realization of the APSEL chips withthe ST 130 nm DNW process, with very encour-aging results, the Italians collaborators involvedin the CMOS MAPS R&D are now evaluatingthe possibility to improve MAPS performancewith the use of modern vertical integration tech-nologies [9]. A first step in this direction hasbeen the realization of a two-tier DNW MAPSby face to face bonding of two 130 µm CMOSwafer in the Chartered/Tezzaron process. Hav-ing the sensor and the analog part of the pixelcell in one tier and the digital part in the sec-ond tier can improve significantly the efficiencyof the CMOS sensor and allow a more complexin-pixel logic. The first submission of verticallyintegrated DNW MAPS, now in fabrication, in-cludes a 3D version of a 8x32 MAPS matrix withthe same sparsified readout implemented in theAPSEL chips. A new submission is foreseen inAutumn 2010 with a new generation of the 3DMAPS implementing a faster readout architec-ture under development, which is still data pushbut could be quite easily evolve toward a trig-gered architecture.

The development of a thin mechanical sup-port structure with integrated cooling for thepixel module is continuing with promizing re-sults. Prototypes with light carbon fiber mi-crochannels for the coolant fluid (total materialdown to 0.15% X0) have been produced andtested and are able to evacuate specific powerup to 1.5W/cm2 mantaining the pixel moduletemperature within the requirements. Thesesupports could be used for hybrid pixel as forMAPS sensors.

References

[1] The SuperB Conceptual Design Report,INFN/AE-07/02, SLAC-R-856, LAL 07-15, Available online at: http://www.pi.infn.it/SuperB

[2] FastSim ref, Available online at: http://www.pi.infn.it/SuperB

[3] V. Re et al., IEEE Trans. Nucl. Sci. 53,2470 (2006).

[4] SLIM5 Collaboration - Silicon detectorswith Low Interaction with Material, http://www.pi.infn.it/slim5/

[5] G. Rizzo for the SLIM5 Collaboration.,”Development of Deep N-Well MAPS in a130 nm CMOS Technology and Beam TestResults on a 4k-Pixel Matrix with Digi-tal Sparsified Readout”’, 2008 IEEE Nu-clear Science Symposium, Dresden, Ger-many, 19-25 October, 2008

[6] A. Gabrielli for the SLIM5 Collabora-tion, ”Development of a triple well CMOSMAPS device with in-pixel signal process-ing and sparsified readout capability” Nucl.Instrum. Meth. A 581 (2007) 303.

[7] M. Villa for the SLIM5Collaboration,”Beam-Test Results of4k pixel CMOS MAPS and High Resistiv-ity Striplet Detectors equipped with digitalsparsified readout in the Slim5 Low MassSilicon Demonstrator Nucl. Instrum. Meth.A (2010) doi:10.1016/j.nima.2009.10.035

[8] E.Paoloni for the VIPIX collaboration,Beam Test Results of Different Configura-tions of Deep N-well MAPS Matrices Fea-turing in Pixel Full Signal Processing, Pro-ceedings of the XII Conference on Instru-mentation, Vienna 2010. To be publishedin Nucl. Instr. Meth, in Phys. Res. SectionA

[9] R. Yarema, “3D circuit integration for ver-tex and other detectors”, Proceedings 16th


http://www.pi.infn.it/SuperBhttp://www.pi.infn.it/SuperBhttp://www.pi.infn.it/SuperBhttp://www.pi.infn.it/SuperBhttp://www.pi.infn.it/slim5/http://www.pi.infn.it/slim5/

19

International Workshop on Vertex Detec-tors (VERTEX2007), Lake Placid (NY,USA), September 23 - 28, 2007, Proceed-ings of Science PoS(Vertex 2007)017.

[10] F.Bosi and M. Massa, “Development andExperimental Characterization of Proto-types for Low Material Budget SupportStructure and Cooling of Silicon PixelDetectors, Based on Microchannel Tech-nology” Nucl. Instrum. Meth. A (2010)doi:10.1016/j.nima.2009.10.138

[11] J. A. Ballin et al., “Monolithic Active PixelSensors (MAPS) in a quadruple well tech-nology for nearly 100% fill factor and fullCMOS pixels,”, Sensors 8 (2008) 5336.

[12] N. K. Watson et al., “A MAPS-based read-out of an electromagnetic calorimeter forthe ILC,” J. Phys. Conf. Ser. 110 (2008)092035.

[13] J. P. Crooks et al., “A monolithic activepixel sensor for a tera-pixel ECAL at theILC,” CERN-2008-008.

4 Drift Chamber

The SuperB drift chamber provides the chargedparticle momentum measurements and mea-surements of ionization energy loss used for par-ticle identification. This is the only device inSuperBto measure velocities of particles havingmomenta below approximately 700 MeV/c. Itsdesign is based on that of BABAR, which has 40layers of centimetre-sized cells strung approxi-mately parallel to the beam line [1]. A subset oflayers are strung at a small stereo angle in or-der to provide measurements along z, the beamaxis.

The drift chamber is required to provide mo-mentum measurements with the same preci-sion as the BABAR drift chamber (approximately0.4% for tracks with a transverse momentum of1 GeV/c), and like BABAR uses a helium-basedgas mixture in order to minimize measurementdegradation from multiple scattering. The chal-lenge is to achieve comparable or better perfor-mance than BABAR but in a high luminosity en-vironment. Both physics and background rateswill be significantly higher than in BABAR and asa consequence the system is required to accom-modate the 100-fold increase in trigger rate andluminosity-related backgrounds primarily com-posed of radiative Bhabhas and electron-pairbackgrounds from two-photon processes. How-ever, the beam current related backgrounds willonly be modestly higher than in BABAR. Thenature and spatial distributions of these back-grounds dictate the overall geometry of the driftchamber.

The ionization loss measurement is requiredto be at least as sensitive to particle discrimi-nation as BABAR which has a dE/dx resolutionof 7.5%. In BABAR, conventional dE/dx driftchamber methods were used in which the to-tal charge deposited on each sense wire was av-eraged after removing the highest 20% of themeasurements as a means of controlling Landaufluctations. In addition to this conventional ap-proach, the SuperB drift chamber group is ex-ploring a cluster counting option which, in prin-


20 4 Drift Chamber

ciple, can improve the dE/dx resolution by ap-proximately a factor of two. This technique in-volves counting individual clusters of electronsreleased in the gas ionization process. In sodoing, we remove the sensitivity of the specificenergy loss measurement to fluctuations in thenumbers of electrons produced in each cluster,fluctuations which significantly limit the intrin-sic resolution of conventional dE/dx measure-ments. As no experiment has employed clustercounting , this is very much a detector researchand development project but one which poten-tially yields significant physics payoff at SuperB.

4.1 Backgrounds

The dominant source of background in the Su-perBDCH is expected to be radiative Bhabhascattering. Photons radiated collinearly tothe initial e− or e+ direction can bring thebeams off-orbit and ultimately produce show-ers on the machine optic elements. This processcan happen meters away from the interactionpoint and the hits are in general uniformly dis-tributed over the drift chamber volume. Large-angle e+e− → e+e−(γ) scattering has the well-known 1/ϑ4 cross section; simulation studies arepresently underway to evaluate the need to de-sign tapered endcaps (either conical or a withstepped shape) at small radii to keep under con-trol the occupancy in the very forward regionof the detector. The actual occupancy and itsgeometrical distribution in the detector dependon the details of the machine elements, on theamount and placement of shields, on the driftchamber geometry, and on the time needed tocollect the signal in the detector. Preliminaryresults obtained with GEANT4 simulations in-dicate that in a 1µs time window at nominalluminosity (1036 cm−2s−1) the occupancy aver-aged over the whole drift chamber volume is3.5 %, and slightly larger (about 5 %) in the in-ner layers. Intense work is presently underwayto validate these results and study their depen-dence on relevant parameters.

4.2 Mechanical Structure

The drift chamber mechanical structure mustsustain the wire load – about 3 tons for 10 000cells – with small deformations, while at thesame time offer minimum material to the sur-rounding detectors. Carbon Fiber-resin com-posites have high elastic modulus and low den-sity, thus offering performances superior toAluminum-alloys based structures. Endplateswith curved geometry can further reduce mate-rial thickness with respect to flat endplates fora given deformation under load. For example,the KLOE drift chamber [2] features 8 mm thickCarbon Fiber spherical endplates of 4 m diam-eter. Preliminary design of Carbon Fiber end-plates for SuperB indicate that adequate stiff-ness (≤ 1 mm maximum deformation) can beobtained with 5 mm thick spherical endplates,corresponding to 0.02X0 (compare 0.13X0 forthe BABAR Aluminum DCH endplates).

Figure 13 shows two possible endcap layouts,respectively with spherical (a) or stepped (b)endplates. We are also considering a con-vex spherical endplate, which provides a bet-ter match to the geometry of the forward PIDand calorimeter systems, and would reduce theimpact of the endplate material on the per-formance of these detectors, at the cost ofgreater sensitivity to the wide-angle Bhabhabackground.

4.3 Drift Chamber Geometry

The SuperB drift chamber will have a cylin-drical geometry. The dimensions are being re-optimized through detailed simulation studiesrespect to BABAR since:a) in SuperB there will be no support tube;b) the possibility is being considered to add aPID device between the drift chamber and theforward calorimeter, and an EMC in the back-ward direction.

Simulation studies performed on several sig-nal samples with both high (e.g. B → π+π−),and medium-low (e.g.B → D∗K) momentumtracks indicate that:a) due to the increased lever arm, momentum


4.4 Gas Mixture 21

(a) Spherical endplates design. (b) Stepped endplates design.

Figure 13: Two possible SuperB DCH layouts.

resolution improves as the minimum drift cham-ber radius Rmin decreases, see Fig. 14; Rmin isactually limited by mechanical integration con-straints with the cooling system and the SVT.b) The momentum and especially the dE/dxresolution for tracks going in the forward orbackward directions are clearly affected by thechange in number of measuring samples whenthe chamber length is varied of 10−30 cm. How-ever the fraction of such tracks is so small thatthe overall effect is negligible.

Figure 14: Track momentum resolution for dif-ferent values of the drift chamber in-ner radius.

The drift chamber outer radius is constrainedto 809 mm by the DIRC quartz bars. As dis-

cussed before, the DCH inner radius will be assmall as possible: since conclusive designs ofthe final focus cooling system are not availableyet, in Fig.13 the the nominal BABAR DCH in-ner radius of 236 mm has been used. Similarly,a nominal chamber length of 2764 mm at theouter endplate radius is used in Fig.13: as men-tioned above, this dimension has not been fixedyet, since it depends on the presence and thedetails of forward PID and backward EMC sys-tems, still being discussed. Finally, as the restof the detector, the drift chamber is shifted bythe nominal BABAR offset (367 mm) with respectto the interaction point.

4.4 Gas Mixture

The gas mixture for SuperBshould satisfy therequirements which already concurred to thedefinition of the BABAR drift chamber gas mix-ture (80%He − 20%iC4H10), i.e. low density,small diffusion coefficient and Lorentz angle, lowsensitivity to photons with E ∼ 10 keV. Tomatch the more stringent requirements on oc-cupancy rates of SuperB, it could be useful toselect a gas mixture with a larger drift veloc-ity in order to reduce ion collection times andso the probability of hits overlapping from unre-lated events. The cluster counting option wouldinstead call for a gas with low drift velocity andprimary ionization.


22 4 Drift Chamber

4.5 Cell Design and Layout

The baseline design for the drift chamber em-ploys small rectangular cells arranged in concen-tric layers about the axis of the chamber whichis approximately aligned with the beam direc-tion. The precise cell dimensions and numberof layers are to be determine for the TDR phasebut the expectation is that they will be between10 and 20 mm on a side and that there will beapproximately the same number of layers as inBABAR (40) if the inner radius is not decreased.The cells are grouped radially into superlayerswith the inner and outer superlayers parallel tothe chamber axis (axial). In BABAR the chamberalso had stereo layers in which the superlayersare oriented at a small “stereo” angle relative tothe axis in order to provide the z-coordinates ofthe track hits. The stereo layer layout in SuperBis to be determined for the TDR and dependson the cell occupancy associated with machinebackgrounds.

Each cell has one 20µm diameter gold coatedsense wire surrounded by a rectangular grid ofeight field wires. The sense wires will be ten-sioned with a value consistent with electrostaticstability and with the yield strength of the wire.The baseline calls for a gas gain of approxi-mately 5 × 104 which requires a voltage of ap-proximately +2 kV to be applied to the sensewires with the field wires held at ground.

The field wires are aluminum with a diameterwhich will be chosen to keep the electric field onthe wire surface below 20 kV/cm as a means ofsuppressing the Malter effect. These wires willbe tensioned in order to provide a gravitationalsag that matches that of the sense wires.

At a radius inside the inner most superlayerthe chamber has an additional layer of axiallystrung guard wires which serve to electrostati-cally contain very low momentum electrons pro-duced from background particles showering inthe DCH inner cylinder and SVT. A similarlymotivated layer will be considered at the outermost radius to contain machine background re-lated backsplash from detector material just be-yond the outer superlayer.

4.6 R&D work

Various R&D programs are underway towardsthe definition of an optimal drift chamber forSuperB, in particular: make precision measure-ments of fundamental parameters (drift veloc-ity, diffusion coefficient, Lorentz angle) of po-tentially useful gas mixtures; study with smalldrift chamber prototypes and simulations theproperties of different gas mixtures and cell lay-outs; verify the potential and feasibility of thecluster counting option.

A precision tracker made of 3 cm diameterAluminum tubes operating in limited streamermode with a single tube spatial resolution ofaround 100µm has been set up.

A small prototype with a cell structure re-sembling the one used in the BABAR DCH hasbeen also built and commissioned. Tracker andprototype have been collecting cosmic ray datasince October 2009. Tracks can be extrapolatedin the DCH prototype with a precision of 80µmor better. Different gas mixtures have beentried in the prototype: starting with the originalBABAR mixture (80%He− 20%iC4H10) used asa calibration point, both different quencher pro-portions and different quenchers (e.g methane)have been tested in order to explore the phasespace leading to lighter and possibly faster oper-ating gas. Fig. 15a shows the space-time correla-tion for one prototype cell: as mentioned before,the cell structure is such as to mimic the overallstructure of the BABAR DCH. Preliminary anal-ysis shows that the spatial resolution are con-sistent with what has been obtained with theoriginal BABAR DCH. A space to time relation isdepicted in Fig. 15b with a 52%He− 48%iCH4gas mixture. This gas is roughly a factor twofaster and 50% lighter than the original BABARmix: preliminary analysis shows space resolu-tion performances comparable to the originalmix, however detailed studies of the Lorentz an-gle have to be carried out in order to considerthis mixture as a viable alternative.

To improve performances of the gas trackera possible road could be the use of the Clus-ter Counting method. If the individual ion-ization cluster can be detected with high ef-


References 23

[cm]-1.5 -1 -0.5 0 0.5 1 1.5

[ns]

0

100

200

300

400

500

600

10H4Space-time relation - 80%He20%C 10H4Space-time relation - 80%He20%C

(a) 80%He-20%iC4H10 gas mixture.

[cm]-1.5 -1 -0.5 0 0.5 1 1.5

[ns]

0

100

200

300

400

500

600

4Space-time relation - 52%He48%CH

4Space-time relation - 52%He48%CH

(b) 52%He-48%iCH4 gas mixture.

Figure 15: Examples of measured space-time relation in different He-based gas mixtures.

ficiency, it could in principle be possible tomeasure the track specific ionization by count-ing the clusters themselves, providing a two-fold improvement in the resolution compared tothe traditional truncated mean method. Hav-ing many independent time measurements ina single cell, the spatial accuracy could alsoin principle be improved substantially. Thesepromises of exceptional energy and spatial reso-lution must however fit with the available datatransfer bandwidth, require a gas mixture withwell-separated clusters and high detection effi-ciency. The preamplifier rise time and noise arealso issues.

Comparisons of the traditional methods toextract spatial position and energy losses andthe cluster counting method are being setup atthe moment of writing the present report.

References

[1] The BABAR Collaboration, The BABAR De-tector, Nucl. Instr. Meth, in Phys Res.A479 (2002) 1.

[2] M.Adinolfi et al., The KLOE Collabora-tion, The tracking detector of the KLOE

experiment, Nucl. Instr. Meth, in Phys Res.b A488 (2002) 51.


24 5 Particle Identification

5 Particle Identification

5.1 Detector concept

The DIRC (Detector of Internally ReflectedCherenkov light) [1] is an example of innova-tive detector technology that has been crucialto the performance of the BABAR first-class sci-ence program. Excellent flavor tagging will con-tinue to be essential for the program of physicsanticipated at SuperB, and the gold standardof particle identification in this energy region isagreed to be that provided by internally reflect-ing ring-imaging devices (the DIRC class of ringimaging detectors). The challenge for SuperB isto retain (or even improve) the outstanding per-formance attained by the BABAR DIRC [2], whilealso gaining an essential factor of 100 in back-ground rejection to deal with the much higherluminosity.

We are planning to build a new Cherenkovring imaging detector for the SuperB barrel,called the Focusing DIRC, or FDIRC. It willuse the existing BABAR bar boxes and mechan-ical support structure. We will attach to thisstructure a new photon “camera”, which will beoptically