current topics in d˜ b-physics arthur...

Current Topics in DØ B-Physics

Arthur Maciel( Northern Illinois University, DeKalb, IL.)

• DØ and the Tevatron

• Muon Triggers, Yields

• The DØ Bs Program

• FCNC, Rare Decays

• DØ RunIIb Upgrades

Assisi, Italy June 20, 2005

FERMILAB-CONF-05-388-E


Tevatron RunII

Tevatron Now

∼ 65-70 pb−1/month delivered !

• Reaching RunII target luminosity

– Peak lums above 1.2 · 1032cm−2s−1

• Pbar recycler up and running

• e-cooling being commissioned

Arthur Maciel(NIU), Beauty’05(20-06-05) 1


Tevatron

Delivered: 1 fb−1

Recorded: 0.8 fb−1

(each experiment)120 GeV p

_

_

F0

A0

E0 C0_

_

B0

D0

_

P1

A1

P8

P3

P2

NS

W

E

RunIIb expecting > 1 fb−1/yr.

(starts ∼Jan. 2006)



DØ RunII

• Data taking efficiencyin 2005 is ∼88%

– currently translates

to 55-60 pb−1/month

• RunIIb upgrades

– Trigger and DAQ

– SMT Layer-0

• Tev B-program suc-

cess largely dependent

on trigger strategies

=⇒

• σ(pp→ bb) ≈ 150µb at 2 TeV

• σ(e+e− → bb) ≈ 7 nb at M(Z)

• σ(e+e− → bb) ≈ 1 nb at Υ4s

• but... σ(bb)/σinel ≈ 10−3



The DØ Detector – RunIIa

Calorimeter

Shielding

Toroid

Muon Chambers

Muon Scintillators

η = 0 η = 1

η = 2

[m]

η = 3

–10 –5 0 5 10

–5

0

5

Keys to the B-program:

Central magnetic

tracking volume:

• Is a RunII addition

• Compact (r < 52cm)

• Modest p resolution

Γµµ = 60 MeV at J/ψ

Clean muon-ID:

• Efficient muon triggers

• single-µ 60% pure at L1

Wide angle coverage:

• Muon chambers; |η| < 2.0

• Tracking volume; |η| < 3.0



The DØ Tracking Volume

Solenoid

Preshower

Fiber Tracker

Silicon Tracker

η = 0 η = 1

η = 2

[m]

η = 3

–0.5 0.0–1.5 0.5 1.0 1.5–1.0

–0.5

0.0

0.5

1.2 m

2 Tesla (SC)

SMT 2ndry

vtx resoln:

• 40µm (r, φ)

• 80µm (r, z)

• Impact parameter resolution

– of ∼ 50µm at pT of 1 GeV

– improving to ∼ 10µm at high pT

10

10

20

30

D0 data (preliminary)

D0 MC single muons40

50

60

σ∝

(d )

,m

0

10 10p , GeV

2

T

Track DCA resolution

(µm) versus pT (GeV)



Muon Triggers

• Keys to the DØ B-Program (and access to unbiased lifetimes)

• Levels 1 and 2 dominated by muon hits, aided by tracking

• Level 3 flexible and fast reconstruction of full event

• Typical rates (Hz):input L1 L2 L3

DØ total 1 MHz 2000 800 50

dimuons 75 20 2

Level 1(hardware)

Level 2(hybrid)

Level 3(software)

muon scint hits muon tracking (indept.) accurate tracking

muon wire hits choice of time gates and matching

central tracks flexible track matching prim. vertex z

prim. vertex z impact parameter impact parameter

see talk Wed.11:20

by Sasha Caroninvariant mass

(In red, tools not yet in use, are “on call” for the higher lum.)Arthur Maciel(NIU), Beauty’05(20-06-05) 6


Examples of Reconstruction Yields

• From J/ψ triggers;

( 5K J/ψ’s / pb−1)

• From single muon triggers;

( 1M events / pb−1)

decay ∼ N/pb−1

B±u −→ J/ψ +K± 21

B0d −→ J/ψ +K∗ (Kπ) 8

B0d −→ J/ψ +K0

s (ππ) 2

B0s −→ J/ψ + φ (KK) 2

Λ0b −→ J/ψ + Λ0 (pπ) 0.3

D̄0µ+νX ( 500/pb−1)

(semileptonic) containing;

B+u → D̄0µ+ν , and

B0d → D∗−(2010)µ+ν =⇒

)2 (GeV/c+π-K - M+π +π-K M0.14 0.16 0.18 0.2

)2 E

ven

ts /

(1.0

0 M

eV/c

0

2000

4000

6000

Run II∅D

preliminary

-1 data - 200 pbµsingle-

173± = 20133 µ+*DN

∆m(Kππ − Kπ) The D∗ signal

with assoc. µ

100/pb−1



The DØ B-Physics Roadmap• Mixing, oscillations (Tania Moulik, today 14:30)

– benchmark tests and measurements completed (∆md)

– extraction of tagging efficiency and dilution from real data

– a first shot at Bs mixing – tool development and a limit

• Hadron lifetimes (A. Sanchez (∆Γs/Γs), Tues. 12:00)

– competitive measurements of lifetimes and their ratios(

τ(Λ0b)

τ(B0d)

,τ(B0

s )

τ(B0d)

)

exclus(J/ψ)

(

τ(B+u )

τ(B0d)

, τ(Bs)

)

semilept

– currently a world’s most precise measurement of τ(Bs)

• Rare decays (S.F. and S.D. -CDF- Tues. morning)

– searches for FCNC decays in both b and c sectors

• HF production and spectroscopy (DB,WW,ECB,UK - Fri. morning)

– bottomonium, b-baryons, excited mesons, X(3872), Bc · · ·Arthur Maciel(NIU), Beauty’05(20-06-05) 8


The DØ Bs Program

• Access to a large sampleof semileptonic Bs decays

– will strongly contribute to

the extraction of ∆(Γs)/Γs

when combined with Bs →

J/ψφ measurements

• Reconstructed Ds mesons

with a nearby muonmode yield/pb−1

D−s

→ φπ− 29 evts.

D−s

→ K∗0K− 25 evts.

1.7 1.75 1.8 1.85 1.9 1.95 2 2.05 2.1 2.15

2000

2500

3000

3500

4000

4500

5000

5500

277±: 13339sD

292±: 5019±D

Mass(φπ) [GeV/c ]2

DØ Preliminary

with assoc.muon, 460 pb−1

M(φπ)

)2

(GeV/c-)K-π+(K M1.6 1.8 2 2.2 2.4

)2 E

vent

s / (

20.0

MeV

/c

0

1000

2000

3000

4000

Run II∅D Preliminary

-1200 pb

< 0π*QµQ

> 0π*QµQ

)-π + K→*0 + X (K- K*0 K+µ → sB

with assoc.muon, 200 pb−1

4933

±376



The DØ Bs Program

• Bs lifetimes: the single

exponential description

(flavor specific)

=⇒

] 2

) [GeV/c-πφMass(

1.75 1.8 1.85 1.9 1.95 2 2.05 2.1 2.15 2.2

2C

and

idat

es /

10 M

eV/c

0

1000

2000

3000

4000

5000

6000

] 2

) [GeV/c-πφMass(

1.75 1.8 1.85 1.9 1.95 2 2.05 2.1 2.15 2.2

2C

and

idat

es /

10 M

eV/c

0

1000

2000

3000

4000

5000

6000(a)

/dof = 1.332χ

DØ Run II Preliminary

Pseudo Proper Decay Length (cm)-0.4 -0.3 -0.2 -0.1 -0 0.1 0.2 0.3 0.4 0.5 0.6

Eve

nts

/ (

0.00

5 cm

)

-110

1

10

210

310

Pseudo Proper Decay Length (cm)-0.4 -0.3 -0.2 -0.1 -0 0.1 0.2 0.3 0.4 0.5 0.6

Eve

nts

/ (

0.00

5 cm

)

-110

1

10

210

310

/dof = 1.092χ


M(φπ)

world’s single

most precise

measurement

of τ(Bs)

τ(B0s) = 1.420 ± 0.043 ± 0.057 ps

∼ 400pb−1

N(µ+D−s

) = 5153 ± 265 ± 450

*** ***



Bs Lifetimes – ∆Γs/Γs

• Bs lifetimes: the double

exponential description

– uses Bs → J/ψφ =⇒

– flavor non-specific final state

– assuming no CP violation,

– discriminate eigenstates

(CP even) BHs mH

(CP odd) BLs mL

• ∆ms = mH − mL significantly

larger than ∆md

• Mass (CP) eigenstates may end up

with a sizeable lifetime difference ∆Γs

Mass (GeV)

5 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.82

Can

did

ates

per

20.

0 M

eV/c

0

50

100

150

200

250

300

350

400 Fit prob: 19.8 %

φ ψ J/→ sB

D Preliminary

450pb−1

483 ± 32

∆Γs = Γ(BHs )−Γ(BLs )

Γ̄ = (ΓH + ΓL)/2

τ̄ = 1/Γ̄



Bs ( ∆Γ/Γ ) – cont.

Bs

B̄s

−→ J/ψφ

Bd

B̄d

−→ J/ψKs

⇐= similar =⇒

untagged decays via

W−

c̄

s

b c

y

x

z

φ

θ

φ

J/ψ

µ+

K

K−

µ−

frameψ rest J/

+

x−y plane

• Final states are common to (mixed) meson & anti-meson.

• Separate the two CP com-

ponents of the decay by

means of polarization prop-

erties (“transversity”) of the

final state.

transversity = cos θ

P→VV; J=0 ⇒ L=0,1,2

S&D waves are CP-even for ψφ

P wave is CP-odd for ψφ



Bs ( ∆Γ/Γ ) – Results

ct [cm]-0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3

mµC

and

idat

es p

er 5

0.0

10-2

10-1

1

10

102

103 DataTotal FitTotal SignalCP-evenCP-oddBackground

φ ψ J/→ sBMass 5.26 - 5.42 GeV

D Preliminary

Transversity-1 -0.8 -0.6 -0.4 -0.2 -0 0.2 0.4 0.6 0.8 1

Eve

nts

per

0.2

0

20

40

60

80

100DataTotal FitCP-evenCP-oddTotal SignalBackground

Fit prob: 61.6 %

φ ψ J/→ sB

) <5.42 GeVs5.26< M(B

(ct) > 5σct/

Preliminary∅D

• A simultaneous fit (unbinned LL)

– to mass,

– proper time evolution,

– transversity distribution,

separates the two decay modes

τ̄ 1.39+0.13

−0.14± 0.08 ps

τL 1.23+0.16

−0.13(s+ s)

τH 1.52+0.39

−0.43(s+ s)

∆Γ/Γ̄ 0.21+0.27

−0.40± 0.20

f(t=0)odd 0.17 ± 0.10 ± 0.02



Bs ( ∆Γ/Γ ) – Results

0.0368 0.0396 0.0424 0.0452 0.048-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

ψ & φ , TθSignal eff. vs

Default fit

Signal mass model

Procedure bias + Detector alignment +Background lifetime Model{

D Preliminary

Γ / Γ ∆

(cm)τc

, cmτc0.034 0.0368 0.0396 0.0424 0.0452 0.048

Γ / Γ

∆

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1 (stat only)σ1

CDF∅D

bandσtheor. 1

WA(f.s. ) σ 1 ±

stat)σwith WA constraint (1 (stat + syst)σ1

Preliminary∅D1σ contours

• The semileptonic (f.s.) τs

measurements provide an

independent relation (or

constraint) between cτ̄

and ∆Γ/Γ̄ ⇓

• See talk by A. Sanchez

(Tues. 12:00) for details

and a discussion on com-

bined results and aver-

ages



Bs Mixing

0 0.02 0.04 0.06

As

ym

me

try

Visible Proper Decay Length (cm)

-1

-0.5

0

0.5

1DØ Run II Preliminary

)-1

(pssm∆A

mp

litu

de

-2.5

-2

-1.5

-1

-0.5

0 2 4 6 8 10

0.5

1

1.5

2

2.5

(stat.)σ 1.645 ±data

σ 1.645 ±data

σ 1 ±data

(stat. only)-1

5.2ps -1

95% CL limit: 5.0ps

(stat. only)-1

5.1ps -1

sensitivity: 4.6ps

• Developing tools to measure ∆ms

— lots of room for improvement

• Currently using only one semilept.

channel: Bs → Ds(φ(KK) π) µ X

and binned likelihoods

• Flavor tagger builds a (b, b) likelihood

from a choice of OS(µ) discrim. vars;

µ , prelT , jetQ , svtQ . . .

• (460 pb−1) 380 tagged events

in VPDL range (-100,600) µm

• Fit inputs;

– sample composition,

– K-factors, VPDL resolution,

– efficiencies, and tagger dilution

∆ms > 5.0 ps−1

at 95% CL

using “amplitude” method



Searches for FCNC B-Decays

• Forbidden at tree-level in SM, must proceed

via higher order WI processes =⇒

• Physics beyond SM becomes potentially

dominant over (competitive with) SM itself

– tree level contribs in Rp-viol. models

– BR’s increase like (tanβ)6 in MSSM

• DØ studies two channels:

s, d W µ

b W µ

t νµ

s, dt

µ

bt

µ

WZ

+ · · ·

Bs −→ µ+µ−

• Cleaner theory predictions

SM: BR = (3.4 ± 0.5) · 10−9

• Blind analyses, normalized

to well measured channels

Bs −→ µ+µ−φ

• Add a spectator strangeness

to LHS of diagrams above

• Larger hadronic uncertainties

(∼30%) SM: BR ≈ 1.6 · 10−6



Results for FCNC B-Decays

• Bs → µ+µ− summary table: BR < ( at 95% C.L.)

published

CDF-I (100 pb−1) 2.6 · 10−6 (PRD57, 1998)

CDF-II (170 pb−1) 7.5 · 10−7 (PRL93, 2004)

DØ-II (240 pb−1) 5.0 · 10−7 (PRL94, 2005)

updates – this conference

DØ-II (300 pb−1) 3.7 · 10−7 (preliminary)

CDF-II (360 pb−1) 2.0 · 10−7 (preliminary)

• Bs → µ+µ−φ

– Presently only one (95% C.L.) limit, CDF-I

BR < 6.7 · 10−5

– DØ signal region still blinded; box to be opened soon.

Sensitivity study shows expected (95% C.L.) limit

BR < 10−5

combined

results

in progress



Towards FCNC D-Decays

• Ideal (and largely untapped) testing ground for new phenomena.

• Attention to FCNC is mainly focused at the I3 = −1/2 sector,

– B-decays; b→ s `+`− , (b→ sγ) , K-decays; s→ d `+`−

– SM: (b→ s) transitions have top-assisted higher order loops

• In contrast, the I3 = +1/2 sector SM transitions involve loops with

the “all-light” d-sector fermions (⇒ an efficient GIM cancellation).

• Consequence: SM expectations for D − D̄ mixing and FCNC D-decays

are very small. However... SM extensions may upset this cancellation,

– (c→ u) transitions induced by new phenomena may exceed

SM expectations by many orders of magnitude.

– see hep-ph/0112235 for a discussion of different scenarios.

∼anything non-resonant that is observed ⇒ New Phen!!

• DØ plan: to set model indept. limits on diff. decay rate dΓ(µ+µ−π)

dm(µ+µ−)



Towards FCNC D-Decays• General strategy:

– Step-1; understand the resonant regions in µ+µ−π spectrum

(long-distance contributions, non-perturbative and therefore

non-calculable).

– Step-2; search for excesses in

continuum regions of spectrum.

• Current status: (∼ 500 pb−1)

First observation of the decay

D±s → φπ±, φ → µ+µ− ⇒

(7σ : 33 ± 7 candidates)

• Spin-off; a (preliminary)

best limit on the decay

D± → φ(µ+µ−) π±

BR < 3.14 · 10−6 (90% CL)

D→φπ+→π+µ+µ− D∅ Preliminary

m(µ+µ−π+) (GeV/c2)

Eve

nts

/ 20

MeV

/c2 d > 0.75d > 0.9

0.96 < m(µ+µ−) < 1.06 GeV/c2



And we have not talked about...

• production

• baryons

• lifetimes and ratios

• spectroscopy

B∗∗ , D∗∗ , Bc

• · · ·

the Υ cross section

Phys. Rev.Lett. 94, 232001 (2005)

(GeV/c)Tp0 2 4 6 8 10 12 14 16 18 20

-1 (

GeV

/c)

TO

Tσ

dy)

/T

/dp

σ2(d

-310

-210

-110

(s) = 1.96 TeV√ |y|<1.8 ∅D(s) = 1.8 TeV√CDF |y|<0.4



RunIIb Upgrades (B-specific)

• SMT Layer-0

– addition of an inner layer: R1 = 26mm⇒ R0 = 17mm

– offsets expected degradation of (non-upgraded) SMT

– 48 modules: 6φ+8z segments, 256 channels each

– installation during next shutdown, Fall’05

– simulation ⇒ 25% improvement in decay

length resolution (and proper time for

hadronic modes)

• Bandwidth increase

– unprescaled reach to lower pT muons

– DØ muons not rate-limited at L1 and L2

– current limiting rate is L3-to-tape at 50 Hz

– proposal: an extra 50 Hz dedicated to B-physics

– basically just increased cpu resources – Fall’05



Impact on ∆ms Program

• Both cases using 1µ-inclusive triggers

• Impact of Layer-0 on semilept mode

not as significant due to ν-smearing

+ 50 Hz Upgrade

10 12 14 16 18 20 22 24 26 28 30

Inte

gra

ted

Lu

mi fo

r O

bse

rva

tio

n a

t 3σ

[pb

]

-1

0

500

1000

1500

2000

2500

3000

3500

4000

CurrentResolution

Semileptonic Mode, B → D µν Xs s

From global fits1σ 95% C.L.

region

CurrentWorld Limit

[ps ]m∆ s-1

10 12 14 16 18 20 22 24 26 28 300

500

1000

1500

2000

2500

3000

3500

4000Hadronic Mode, B → D π s s

Inte

gra

ted L

um

i fo

r O

bserv

ation a

t 3

σ[p

b ]

–1

+ 50 Hz

Upgrade

w/ Layer 0

Current

[ps ]m∆ s-1

From global fits1σ 95% C.L.

region

CurrentWorld Limit

Int. Lum. needed

to achieve a 3σ

measurement as a

function of ∆ms

semileptonic mode ⇒

⇐ hadronic mode

• Both cases assuming single-channel analyses

• Semileptonic mode hits a resolution wall

• Hadronic mode needed for extended reach



Conclusion

• DØ is producing a wealth of B-results.

• A competitive B program, with decisive

impact on world’s averages and limits.

• Key RunII measurements are on track,

with uncertainties dominated by statis-

tics, not systematics.

• B program to be further enhanced by

RunIIb upgrades.



Backup Slides

Extra



Flavor Tagging

• Work in progress: tool development and calibration exercises with

Bu and Bd semileptonic decays to (D̄0 µ+ ν X) final states.

• The (D̄0 µ+) sample is divided into two mutually exclusive com-

ponents

– The “neutral” sample (Bd enriched):

D̄0 has an associated pion (opp. charge to muon)

Dominated (∼ 85%) by B0d → D∗−(2010)µ+ν

– The “charged” sample (Bu enriched): the remainder,

Dominated (∼ 85%) by B+u → D̄0µ+ν

– See plot and yields two pages back

• Flavor at decay time given by sign of D̄0-associated muon

• At production time, different tagging algorithms are under tests

OS: soft muon “SLT” jet charge “jetQ” sec.vtx charge “vtxQ”

SS: soft pion “SST” and their combinations



Combined Tagging Performance - I

• Tagged sample divided into two uncorrelated components

SLT: events have OS muon tag. Other taggers not used.

jetQ⊕SST: remaining events. Use combined jetQ and SST;

∗ reject events with conflicting jetQ and SST results

∗ accept events tagged by either jetQ or SST or both

200 pb−1 Efficiency Dilution(%)

(%) charged neutral

SLT 5.0 ± 0.2 44.8 ± 5.1 44.8 ± 5.1

jetQ⊕SST 68.3 ± 0.9 27.9 ± 1.2 14.9 ± 1.5

↔→

∆md = 0.456

±0.034 (stat)

±0.025 ( sys )

ps−1

( consistent with WA )

m)µ (T/pB MxyL0 500 1000 1500

-0.4

-0.2

0

0.2

0.4

m)µ (T/pB MxyL0 500 1000 1500

-0.1

0

0.1

SLT jetQ⊕SST



Combined Tagging Performance - II

• A different (OS-only) tagger used in Bs mixing

• Here used on B±u and B0

d samples as cross checks

• Flavor tagger builds a (b, b) likelihood from a choice of OS(µ)

discriminating variables; µ , prelT , jetQ , svtQ . . .

460 pb−1 Efficiency Dilution(%)

(%) charged neutral

OST 5.0 ± 0.1 46.8 ± 3.0 44.8 ± 4.2

VPDL (cm)

0 0.05 0.1 0.15 0.2 0.25

Asy

mm

etry

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1X Sample

0D +µ


VPDL (cm)

0 0.05 0.1 0.15 0.2 0.25

Asy

mm

etry

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1X Sample*- D+µ


∆md = 0.558

±0.048 (stat)

ps−1

( consistent with WA )B±u

B0d



Bs → D±s1(2536)µνX

D±

s1(2536) is an orbitally

excited (c, s) pair in a

JP = 1+ state

• Reconstruction

18 ± 5 events

significance > 3σ

• Decay chain is

D±s1(2536) → D∗± K0

s

D∗+ → D0π+

D0 → K−π+

• Muon plus 5-Track final state

)2

(GeV/c0Invariant Mass of D* K2.5 2.55 2.6 2.65

Eve

nts

/1.5

MeV

2

4

6

8

10DØ Run II Preliminary

485 pb−1

• Next step: investigate

signal properties



Current ∆ms Reach

0

500

1000

1500

2000

2500

3000

3500

4000

4500

Current DØ

Sensitivity

Current DØ

Limit @ 95% CL

4 5 6 7 8 9 10 11 12[ps ]m∆ s

-1

Inte

gra

ted

Lu

mi fo

r in

dic

ate

d s

en

sitiv

ity

(exp

ecte

d lim

it @

95

% C

.L.)

-1[p

b

]

x ~2: additional channels,

K*K, 3π, K K

x ~2.4: electron flavor tagging

x ~1.7: selection, improved S/N

x ~2: flavor tag probability

event-by-event

x ~1.3: resolution

event-by-event

Further improvement,

unbinned likelihood:

0S

• How current

sensitivity will

scale with

luminosity if

analysis remains

unchanged.

• Expected

improvements

independently

of upgrade.


current topics in d˜ b-physics arthur...

Documents