David Hitlin

1
Physics and Detector Challenges
at a
Super Factory
B
David Hitlin Caltech March 20, 2003
2
Parsing the title of the talk

• Physics Challenges
• The improvement of measurement precision is a
  sufficient motivation for a 1036 machine, if and
  only if the improved precision takes us into
  discovery territory
• There are indeed areas in which large data
  samples (10-50 ab-1) can lead, with reasonable
  certainty, to measurable new physics effects, by
  increasing precision or making certain
  measurements possible
• The context is also important
• What new physics potential exists with a 10-50
  ab-1 sample that doesnt exist with a 0.5-1 ab-1
  sample?
• What can an asymmetric ee- machine at 1036
  contribute beyond what can be done at hadron
  experiments (ATLAS, CMS, LHCb, BTeV)?
• What is the time window for 1036?
• Detector Challenges
• What do we need in a detector to do physics at a
  1036 machine?
• Should there be an upgrade of BABAR, or a totally
  new detector?
• What RD is required on new detector subsystems?

3
The New Physics Bible according to Nir

• CP violation is an excellent probe of new physics
• The Standard Model CKM mechanism has a single
  source of CPV and makes quantitative predictions
• New sources of flavor and CP violation can induce
  large deviations from the Standard Model
  predictions, many of which are not obscured by
  hadronic uncertainties
• Henceforth in this discussion, I will emphasize
  the supersymmetric Standard Model as an example,
  although other extensions of the Standard Model
  can also produce observable effects
• The supersymmetric SM has 124 independent
  parameters, 44 of which are CP-violating
• What are the constraints of existing measurements
  of CPV on SUSY model building?
• What are the prospects that future CPV
  measurements will uncover deviations from the SM
  predictions?
• Having found that ACP in
  agrees with CKM prediction, we are beyond the era
  of seeking alternatives to the CKM phase and must
  now search for new physics by finding loop
  corrections to the CKM picture

4
New CP Violating effects must be there

• CP effects in the flavor sector that are not
  accounted for by the CKM phase must exist
• If they do not exist, SUSY and other models
  constructed with the same motivation will be
  ruled out
• The sensitivity required to see these effects can
  be reached
• It is possible, though not likely, that SUSY
  could be discovered through loop effects before
  there is explicit production of new particles at
  LHC
• Assume that evidence for SUSY is found at the LHC
  or NLC
• What will we actually know?
• The masses of some of the SUSY partners gluino,
  squark, ..
• Something about coupling constants
• Perhaps the identity of the LSP
• Even if the first evidence for SUSY comes from
  LHC, it will be important to study CPV in flavor
  physics at the scale of 1010 to 1011 B decays

5
SUSY mass spectra for the 9 Snowmass points
slopes
Ghodbane and Martyn
6
Many SM extensions yield measurable effects in B
physics
Generic Little Higgs
Little Higgs wMFV UV fix
Generic extra dim w SM in bulk
Extra dim wSM on brane
SUSY GUTs
SupersoftSUSY breakingDirac gauginos
MSSMMFVlarge tanb
MSSMMFVlow tanb
Effective SUSY
SM-like B physics
New Physics in B
data
after G. Hiller
7
Mapping SUSY-breaking schemes to flavor models
Exact Universality
MSUGRA
Approximate Universality
GMSB
No Universality
Approximate CP
AMSB
MFV
GMSB
Extended MFV
SUSY GUTS
? ? ?
? ? ?
J. Hewett
8
Constraints on SUSY from existing measurements

• In order to obey the constraints from K decay
• Indirect CPV in and
  decays e (2.28 ? 0.02) x 10-3
• Direct CPV in decays
  Ree?/e(1.66 ? 0.16) x 10-3
• it is necessary to invoke one or more of the
  following
• Heavy squarks
• Universality
• Alignment
• Approximate CP CPV phases are small
• All viable models of SUSY-breaking use one or
  more of these mechanisms
• Two other measurements
• ACP in decay Im l?K
  0.734 ? 0.054
• Limits on EDMs (through T violation and CPT)
• impose serious additional constraints
• For example, ACP effectively kills Approximate CP
  models
• EDM limits imply that the source of CPV beyond
  the Standard Model in models with minimal flavor
  violation is Yukawa couplings, which can be
  flavor dependent

9
Effects of SUSY breaking on CPV in flavor physics

• Specific models produce specific CPV patterns
• There are a variety of models of SUSY breaking on
  the market
• Many of these models generate specific,
  calculable CP-violating effects in hadronic and
  rare B decays
• Other extensions (extra dimensions, Little
  Higgs,.) have the same sorts of effects,
  although they often have distinguishable patterns
• In order to exploit CP violation as a tool to
  search for physics beyond the Standard Model we
  must do two things
• Achieve the highest meaningful precision on CPV
  (a, b, g ) measurements of the B unitarity
  triangle
• This requires several x 10 ab-1
• Measure kinematic distributions and CP-violating
  (and sometimes CP-conserving) asymmetries in very
  rare decays with branching fractions of lt10-5,
  both inclusive and exclusive
• These are decay modes such as
  where we have at present only a handful of events

10
Probes of new physics - I

• Measure the CP asymmetry in modes other than
  that measure sin2b in the Standard
  Model
• Precision of benchmark sin2b in
  can improve to the ?1 level
• Expect the same value for sin2b in
  ,but different
  SUSY models can produce different asymmetries
• A great deal of luminosity is required to make
  these measurements to meaningful precision

11
From the BABAR Physics Book (SLAC-R-504)
12
Variations from SM predictions can be substantial

• Three examples
• mSUGRA
• SU(5) SUSY GUT with nR
• U(2)

Goto, et al.
13
Many CP asymmetries can be changed by SUSY
Ciuchini, Franco, Martinelli, Masiero,
Silvestrini
14
SUSY models are already constrained by ACP, Dm,
EDM
U(2) model of Masiero, et al. There are two real
parameters, j and y

• ? j -0.25, y 0
• ? j -0.25, y -0.25
• x j -0.5, y -0.25

15
Other Standard Model extensions also change CPV
Correctionto SM prediction EffectiveSUSY Enhanced chromo-magneticdipole SUSYwithoutR-parity
Grossman and Worah
16
An example CP in

• The fact that the CP asymmetry in
  is so close to theStandard Model prediction
  tells us that new CP-violating contributions to
  b?d transitions (via ) are small
• The fact that is close to the
  Standard Model value tells us that the
  helicity-conserving part of is
  small.
• The helicity-changing part of , i.e.,
  and could still be large
• and enter the
  supersymmetric gluonic penguin that contributes
  to
• This produces a series ofinter-related
  constraints

Chang, Masiero, Murayama
Change in B(b?sg) ACP fKS Dms
17
BABAR fKS results
Ncand 66 Purity 50
81.3 fb-1
18
Current ACP(fKs) has large errors, but opposite
sign
Pure CKM forbidden penguin amplitude

• Interesting, but not yet a persuasive
  case for new physics

19
CP violation in modes that measure sin2b
Decay mode B x 10-6 Decay process S C
440 0.734?0.054 l0.95?0.04
29 0.76?0.36 -0.26?0.22
4 -0.39?0.41 0.56?0.43
20 -0.46?0.49 0.31?0.29
100 0.31?0.46 l0.98?0.27
It is certainly premature to draw any conclusions
about disparities. One mode is clean fKS.
Branching ratio is small. Could a statistically
persuasive case for a different ACP from J/yKS
be made?
20
What level of precision is required ?

• Statistical/systematic error on sin2b from
  will improve to somewhat beyond the
  1 level. More than adequate
• SUSY effects on sin2b in other modes can be quite
  large, tens of percent of the CKM value
• With what precision must one measure sin2b in
  other, more difficult decay modes in order to
  establish an effect?
• An example
• sin2b ( ) 0.75 (its current value),
  but the error is reduced to1, s 0.0075,
• sin2b ( ) 0.60, i.e., the SUSY
  contribution to is 20
• For a 5 sigma effect Dsin2b 0.15/5 0.03, a
  5 measurement
• This requires a data sample of the size provided
  by a 1036 asymmetric B Factory

21
Extrapolated statistical errors on CP asymmetries
BABAR measurement errors
10 to 50 ab-1 are required for a meaningful
comparison
Currentprecision
22
Probes of new physics - II

• Measure branching ratios and kinematic
  distributions in rare decays that are sensitive
  to new physics, particularly those involving b?s
  transitions
• Requires tens of ab-1

23
Kinematic distributions and CP asymmetries in
rare decays
In SUGRA, sign of C7 determines sign of AFB

• Bauer, Stech Wirbel
• Ball and Braun
• Melihov, Nikitin and Simula

• SM, gt SUGRA with ?C7,
• ?MIA with suppressed Br, ? MIA with enhanced Br

Ali, et al.
Standard Model predictions are robust
24
Probe of SUSY in and
SUGRA
SM
Standard Model predictions are robust
Ali and Safir
25
CPV in exclusive radiative decays
Ali and Lunghi
26
MSSM CP asymmetry in b ? sg
Bartl, Gajdosik, Lunghi, Masiero, Porod,
Stremnitzer and Vives, hepph/0103324

• No EDM constraint
• Obey EDM constraint

27
The effect of extra dimensions on UT parameters
Buras, et al.
28
Probes of new physics - III

1. Measure sides and angles of the Unitarity
   Triangle to best possible precision

Improve measurements of Vub and
Vcbessentially independent of new
physicsSuper B Factory using the recoil technique
Improve measurement of DmdSuper B Factory
Measure DmsHadron machine
Measure sin2aeffSuper B FactoryHadron machine
Measure sin2aSuper B Factory using p0p0
Measure gSuper B FactoryHadron machine
Measure sin2bSuper B FactoryHadron machine
Improve calculations of Vub, Vcb, Lattice

• Test a b g p to 5-10

29
The B beam technique

• Reconstruct a very large sample of of hadronic
  decays at the Y(4S)
• In 10 ab-1, there are 4 x 107 fully reconstructed
  Bs in which thefour momentum of the recoil is
  known
• Use this sample to study semileptonic decays and
  rare (inclusive) decays
• The B beam technique, unique to ee-, sacrifices
  statistics, but
• Improves kinematics reducing model dependence
  in Vub and Vcb studies
• Reduces background for rare decays, especially
  those involving photons and neutrinos

30
Isolating the penguin contribution to sin2a using
With 10 ab-1, the Gronau-Wyler construction can
place a stringent limit on penguin amplitudes
but there is a 4-fold ambiguity!
sin2aeff 0.020.340.05 with 2aeff 2a 2d
?
s(Da) 4 to 10?
Cahn, Roodman
31
An independent estimate of the Gronau-Wyler
construction
Uses current central values
32
Measuring g with B? DK

• Gronau-Wyler, Atwood, Dunietz and Sonimethod
• Comparison of BRs for B?DK modescan allow
  extraction of g
• There is an 8-fold ambiguity
• With sufficient luminosity, it is possible to
  resolve the ambiguity
• with 10 ab-1, it appears that a precision
  of Dg ?1?-2.5 ? can be achieved
• Study was done with 600 fb-1, scaled to
  10 ab-1

Soffer
Soffer
33
Snowmass 2001 scenario for improvement in the
precision of CKM matrix elements
Eigen, Kronfeld, Mackenzie
34
A projection to 2010 by the CKM Fitter group
35
Improvement of UT measurements can test the SM
(Buras)

• Optimal Unitarity Triangle Test
• Improve measurements of Vub and Vcb, which
  are essentially independent of new physics
  contributions
• This is best done at a 1036 B Factory using the
  B beam technique
• Measure Dms domain of hadron machines
• Improve theory estimates Vub, Vcb, and
• This yields a prediction for g, which is
  measurable both at aB Factory and in hadron
  experiments
• Test of the mixing matrix element
• With the above and with improved knowledge of eK
  and Dmd , we have the best possible prediction of
  the mixing matrix element and thus of mt, which
  can be compared with improved direct measurements

36
Simulations of more of these measurements are
needed

• Calculations needed at 10 and 50 ab-1
• How well can we measureVub with the recoil
  techniquefB with recoil techniqueagAFB in
  s??- or K()??- vs B(B?rg)
• mixingB(t?mg)

PRAVDA Monte Carlo tool with a 1036-capable
detector is nearly ready for these studies
37
Statistics and systematics

• Nearly all important CPV measurements will remain
  statistics limited
• Certain measurements, such as ACP in
  will be systematics limited at below the 1
  level
• Other measurements, such as the extraction of
  sin2a or Vub will be limited by theory
• Many of the most interesting measurements will be
  limited by statistics and backgrounds
• This leads to the question of whether an upgraded
  detector can do better than a
  extrapolation would indicate
• Does improved momentum resolution and improved
  particle ID lead to a better measurement of Spp
  by improving S/B ?
• Does improved photon energy and angular
  resolution lead to a better measurement of tagged
  ?
• Does longitudinal segmentation in the EMC lead to
  better p/e separation and thus better tagging?

38
Comparison of ee- B Factories and hadronic
experiments
Wurthwein
39
What is the future of experimental flavor physics?
ee- experiments Hadron experiments
The current lineup BABAR, Belle CDF, DØ
The situation in 2010 SuperBABAR or SuperBelle ATLAS, CMS, LHCbBTeV

• Total BABAR, Belle data samples will amount to
  800-1000 fb-1 each
• CDF (DØ), in areas which overlap ee-, are being
  calibrated with TeV-II data Würthwien(SSI02)
  for untagged B?hh- 2 fb-1(CDF) ? 500
  fb-1(BABAR, Belle)CDF/DØ can, of course, study
  Bs decay, but is unlikely, in general, to
  markedly improve on ee- results in other areas
• LHC experiments will bring statistics to the next
  level
• In this context, is a new, very high luminosity
  ee- effort warranted?
• 
  
  

40
LHCb physics performance
41
At 1036, ee- is fully competitive in rare decay
studies

Two arm BTeV
SLAC-PUB-8970
42
Comparison of 1 year yields BTev and Super B
Factory
Mode BTeV BTeV Super B Super B
Yield Tagged Yield Tagged
Bs?J/yh(?) 12650 1645 - -
B-?fK- 11000 11000 14000 14000
B0?fKs 2000 200 5000 1500
B0?Kmm- 2530 2530 1000 1000
Bs? mm- 6 0.7 -
B0?mm- 1 0.1 0 -
D?pD0, D0?K-p 108 108 1.6x107 1.6x107
43
A BTeV-generated comparison (updated from 1034)

• Number of flavor tagged B0?p p - (B0.45x10-5)
• Number of B- ? D0 K- (Full product B1.7x10-7)
• Bs , Bc and Lb studies are not done at U(4S) ee-
  machines

44
Comparison of hadronic and 1036 reach

• A comparison from the Snowmass E2 Group summary

45
The 1036 environment
25MHz

• Main concerns
• Machine-related backgrounds
• synchrotron radiation
• particle backgrounds, due primarily to continuous
  injection
• Radiation dose
• Physics backgrounds hadronic split-offs, ..

DIRC
100 Occupancy
7MRad/y
EMC
DCH
SVT
gt10 hits/crystal/event
46
There is an upgrade path from BABAR to SuperBABAR

• If it were feasible to modify the existing BABAR
  detector for use at a 1036 machine, there would
  be substantial savings in time, money and effort
  over a completely new detector
• Upgrading the existing detector is beneficial in
• Reducing costs by reuse of detector components
  and existing IR infrastructure
• Use of existing software as a basis for new
  programs
• Packaging an attractive proposal for funding
  agencies
• An affordable, fast, radiation hard
  electromagnetic calorimeter is the key to the
  morphing of BABAR into SuperBABAR
• An LXe EMC fits into the existing BABAR
  solenoid/flux return
• There is a substantial cost saving over most
  crystals
• Tracking with pixels/strips and a compact readout
  DIRC arecompatible with this design

47
An upgrade path from BABAR to SuperBABAR

1. IFR upgraded(ongoing)
2. Remove SVT,DCH, EMC,DIRC
3. New EMC liquid Xe
4. New tracker Two inner pixellayersSeven(?)
   thindouble-sidedSi-strip archlayers
5. New DIRC(s) with compact readout

SuperBABAR
BABAR
48
The MSSM (Minimal Symbolic Straw Man) Upgrade
Detector

• This BABAR upgrade design has not been optimized
• It is certainly possible to improve upon this
  design
• This will be among the first orders of business
  when the physics foundation for 1036 has been
  solidified
• We are currently implementing this design into a
  fast Monte Carlo called PRAVDA
• Implementation includes the flexibility to
• Vary parameters within a detector subsystem
• Swap technologies for a given subsystem
• TRACKERR package (complete error matrix) for
  vertex/tracking with an all silicon tracker has
  been implemented
• Use parameterized descriptions for PID, EMC and
  IFR
• A shower library is also under development for
  the EMC

49
Vertexing and Tracking

• Pixel layers needed near beampipe
• Double-sided strips for main tracking
• Drift chamber is unlikely to survive 1036
• An all silicon tracker with two pixel
  layers seven double-sided strip layersis a
  good candidate
• Router 60 cm
• It is crucial to have a thin silicon chips and a
  light mounting structure to have adequately small
  multiple coulomb scattering

50
A silicon tracker with adequate momentum
resolution is feasible
Current DCH
Double-sidedstrip _at_ 100mm
A proposal to INFN to develop very thin
double-sided detectors is in preparation
Forti TRACKERR/PRAVDA
51
Particle ID a new kind of DIRC

• Barrel
• Fused silica radiator bars are adequately
  radiation hard
• Background in existing SOB is far too high at
  1036
• New non-SOB DIRC is under development in SLAC
  Group B
• Quartz is sufficiently radiation hard
• Need pixel readout to remove SOB
• Use pixelated PMT ? readout outside the flux
  return
• Endcap
• Requires single photoelectron readout in a
  magnetic field

52
Electromagnetic Calorimeter

• Requirements
• Good energy resolution
• Radiation hardness
• Excellent energy and position resolution
• Large dynamic range
• Uniformity and stability
• Can be met by new crystals LSO, GSO, , which
  are expensive
• Desirable attributes
• Longitudinal segmentation for best possible p/e
  separation
• Minimal interruption in barrel/endcap region
• These are features of a scintillating liquid
  xenon calorimeter, which is under development at
  Caltech

53
Comparison of CsI(Tl), LSO, Liquid Xe
CsI(Tl) LSO LXe
Atomic number Z 54 effective 65 effective 54
Atomic weight A 131
Density (g/cc) 4.53 7.40 2.953
Radiation length (cm) 1.85 1.14 2.87
Molière radius (cm) 3.8 2.3 5.71
l scint (nm) 550 420 175
t scint (ns) 680, 3340 47 4.2, 22, 45
Light yield (photons/MeV) 56,000 (6436) 27,000 75,000
Refractive index 1.8 1.82 1.57
Liquid/gas density ratio 519
Boiling point at 1 atmosphere (?K) 165
Radiation hardness (Mrad) 0.01 100 -
Cost/cc 3.2 gt7 (50 ???) 2.5
54
Unit cell of the LXe EMC

• Hexagonal cells of 1 Molière radius in
  transverse dimension are formed from thin
  quadraphenyl butadiene (TPB)-coated eptfe sheets
• Cells are not load-bearing, thus thin
• Longitudinal segmentation is provided
  byTPB-coated optical separators, with WLS
  fibers sensitive only in a particular segment
• Three segments is probably optimal
• Massless gap ascertain whetherthere was an
  interaction in materialin front of the EMC
• 2, Two larger segments, with divisionnear
  shower max
• Fibers are read out by a pixelized APD,located
  in the LXe volume
• Clear fibers between coil segment and APD
• Redundant readout is simple and inexpensive
• All readout at rear, minimizing nuclear counter
  effect

55
Instrumented Flux Return

• High rate capability
• Good time resolution
• Stable response
• RPCs, LSTs do not have adequate rate
  capability, at least in the end cap region
• Barrel will be upgraded with LSTs in 2004/5
• Endcap may require MINOS-type scintillating
  strips

56
EMC solid angle coverage could be substantially
improved
57
Thr road ahead

• There is discovery potential in 10 to 50 ab-1
  data samples
• Is there an interested community ?
• Yes, drawn from the BABAR and Belle communities
  others
• Can we obtain funding for a 1036 collider and
  detector ?
• Not certain, but possible, given the appropriate
  circumstances
• What is the appropriate time window for a 1036
  machine ?
• It should ideally take data on a time scale
  comparable to the LHC experiments (ATLAS, CMS,
  LHCb) and, potentially BTeV
• The HEPAP-mandated P5 is currently making
  decisions involving major new US facilities
• Its first round includes a decision on BTeV

58
PEP-II luminosity scenario through FY2008
59
PEP-II/Super B Factory Peak and Integrated
Luminosity
60
Technically limited schedule
Super B Factory
PEP-IIBABAR
61
Super B Factory activity at KEK, SLAC

• The Directors of SLAC and KEK have encouraged
  cooperation between BABARians and Bellies on
  future activities, since they believe that
  there will be at most one new high luminosity B
  Factory
• Both Directors agree that high luminosity means
  1036
• The core of a Super B Factory effort will likely
  be drawn from theBABAR and Belle Collaborations
• Accelerator and detector RD is underway in both
  labs and at several of the collaborating
  institutions
• Workshops
• KEK has held four workshops (the most recent on
  February 4)
• There will be a workshop at SLAC on May 8-10 to
  explore in detail opportunities to probe physics
  beyond the Standard Model at a 1036 asymmetric B
  Factory
• There will be an ICFA Accelerator Workshop at
  SLAC in October, focusing on very high luminosity
  ee- circular machines

62
Workshop on the discovery potential of an
asymmetric 1036 machine
SLAC May 8-10
63
The Physics Challenge

• There is a substantial difference in the
  discovery potential of a 0.5-1 ab-1 data sample
  and a 10-50 ab-1 data sample
• With this large data sample we will
• by measuring CP asymmetries and kinematic
  distributions in rare decays
• either find clear SUSY effects and make the first
  measurement of a SUSY phase, or
• place highly constraining limits on SUSY-breaking
  models
• take the various overconstrained tests of the
  Standard Model using the Unitarity Triangle to
  their systematic limit
• 3. produce the most sensitive probes for new
  physics through
• mixing
• searches for lepton flavor violation in t ?mg

64
The Detector Challenge

• In order to do physics at 1036, we must develop
• thin, rad hard pixel and double-sided strip
  detectors and associated readout
• a particle ID system, such as the DIRC with pixel
  PMT readout and no SOB, that can take high rates
• a high quality, fast, rad-hard electromagnetic
  calorimeter
• an IFR that can handle high rates
• an appropriate trigger/DAQ system
• RD is either starting or is underway in many of
  these areas

65
Conclusions

• Detailed studies of CP violation in B meson decay
  (and D and t decay) with samples of 10-50 ab-1
  provide a sensitive probe of to new physics such
  as SUSY
• These studies are vital to an understanding of
  the flavor sector of any extension of the
  Standard Model
• Estimates of physics capabilities of a Super B
  Factory are promising
• Detailed studies of machine and physics
  backgrounds and of limiting systematic errors for
  the experiment are underway
• Capabilities of a Super B Factory are
  complementary to those of hadronic experiments
• Both ee- and hadronic experiments are needed to
  fully explore the realm of flavor physics

66
WANTED
Dead or Alive