Radiative Decays of the X(3872) in BaBar, SLAC Experimental Seminar Bryan Fulsom, March 3, 2009 1 Radiative Decays of the X(3872) in BaBar Bryan Fulsom University of British Columbia SLAC Experimental Seminar March 3, 2009
Jan 12, 2016
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
1
Radiative Decays ofthe X(3872) in BaBar
Bryan Fulsom
University of British Columbia
SLAC Experimental Seminar
March 3, 2009
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
2
Outline• Background Information
– Introduction to Charmonium and the X(3872)
– Recent Experimental Results
– Theoretical Predictions
• Analysis Description– The BaBar Experiment
– Event Reconstruction and Selection
– Fit Strategy and PDFs
• Results– Validation (MC and B c1,2 K)
– Systematic Uncertainties
– X(3872) Results and Implications
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
3
Background Information
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
4
Introduction• November Revolution: discovery of the J/ in 1974
• First evidence of the charm quark
– Strong confirmation of the quark model
• Charmonium is a bound state of cc
• Described by phenomenological models:
– Analogous to hydrogen, positronium
– Theory has enjoyed fairly good success
• Several recent results from the B Factories:
– Some discoveries fit naturally into the charmonium system
– Many others cannot be accommodated by the model
• QCD allows for more “exotic” possibilities
– hybrids, tetraquark states, four-quark molecules
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
5
Basic Theoretical Description• Charmonium potential models (phenomenological):
– non-relativistic (charm quarks are “heavy” compared to binding energy)
– strong force potential via one gluon exchange (similar to Coulomb force)
– quark confinement (increases linearly with separation)
• Typical representation:
plus extensions to include spin-dependent terms, relativistic corrections, etc.
• Many successful predictions on masses, widths, branching fractions
• Lattice QCD has made great advancements, but won’t be discussed here
...3
4)( br
rrV s
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
6
Charmonium Production• Colour-suppressed b c decay
– Predominantly from B cc K
• e+e- annihilation/Initial State Radiation (ISR)
– e+e- collision at/below nominal c.m. energy
– JPC = 1--
• Double charmonium production
– Typically one J/ or (2S), plus second cc state
• Two-photon production
– Access to C = + states
• pp annihilation
– All quantum numbers available
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
7
Charmonium Decays• Annihilation:
– Generally suppressed for bound state
– Decay to leptons is a clean experimental signal
– J/ e+e-, +-, (2S) e+e-, +-
• Radiative:
– EM radiative transition emitting photon
– Emit gluons producing light quarks
– c1,2 J/ (2S) J/+-
• Strong interaction:
– Dominant above ~3.72 GeV (DD threshold)
– OZI-suppressed below this mass
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
8
Exotic QCD• Meson Molecules
– Weakly bound state of two mesons
– Binding at long distance via pion exchange
– Mesons decay as though free
• “Tetraquarks”
– Diquark-antidiquark bound by spin-spin interactions
– Multiplets with non-zero charge and strangeness
– Decay via rearrangement and dissociation
• Charmonium hybrids
– Charmonium with excited gluon
– Doubly degenerate octet of states with exotic JPC
– Decays similar to charmonium, though predictions vary
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
9
Charmonium SpectrumSpectroscopic
notation:
N2S+1LJ
where L=S,P,D,F,...
Angular momentum:
J = L+S
S(qq) = 0 or 1
Parity:
P = (-1)L+1
Charge conjugation
C=(-1)L+S
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
10
The X(3872)
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
11
Discovery of the X(3872)• Belle discovered a signal in B J/+- K; confirmed by CDF, D0 and BaBar
• Narrow (<2.3MeV) particle with mass m(X)=3871.4+/-0.6 MeV/c2
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
12
X(3872) J/ Dipion Mass
• Belle and CDF analysed the mass distribution from X J/
• Both seem to favour a “-like” shape, with J/- in an S-wave
• Shape in BaBar is similar, no attempt to fit
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
13
X(3872) Angular Analyses• CDF analysed angular distribution of daughters,
and tested against various JPC assignments
• Angular analysis compatible with both 1++ and 2-+
– 2 prob.(1++)=27.8%, (2-+)=25.8%
Note: Angular analysis
from Belle favours 1++
disfavours 0++, 0-+, 1+-
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
14
Evidence for X(3872) J/ • Both BaBar and Belle have found evidence for this radiative decay
• J/ and have charge parity C = –, implies X(3872) C = +
)0.4(10)1.06.08.1(
)/)((6
JXXKBBF
)4.3(10)3.00.13.3(
)/)((6
JXXKBBF
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
15
X(3872) D0D0*K• Belle discovered X(3872) in B D0D00K
– Found mass 2.0 higher than W.A. for X(3872)
– Recent update confirms D0D0* decay (8.8)
– Compute m(X)=3872.60.50.4 MeV/c2
• BaBar search confirms X(3872) signal (4.9)
– Ratio of D0D00/D0D0 matches D0* expectation
– Mass ~4.5 above X(3872)
– Angular study inconclusive
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
16
X(3872) Interpretation• Summary of X(3872) Properties:
– Narrow with mass m(X)=3871.4+/-0.6 MeV/c2
– Observed in X(3872) J/, dipion mass is “-like”
– Also seen in decays X(3872) D0D0* and X(3872) J/– Spin-parity identified as either JPC = 1++ or 2-+
• Charmonium Hybrid
– Lightest mass prediction m(ccg)>4.2GeV/c2
• Tetraquark State
– No evidence for charged/neutral mass splitting
– No evidence for charged partners
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
17
X(3872) Interpretation• Conventional Charmonium
– Angular analysis: c1(23P1) (1++) or c2(11D1) (2-+)
– Not expected to violate isospin, X J/
– X(3872) is narrow and does not decay X DD
– Expect c2 J/ to be suppressed
– c1(2P) somewhat inconsistent with predicted mass
Study radiative decays as a diagnostic!
• D0D0* Molecular interpretation
– m(D0) + m(D0*) = 3871.8+/-0.4 MeV/c2
– Decays to X(3872) J/, D0D0*, J/ expected
– Compatible with JPC = 1++ assignment
– Successful predictions vary by model
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
18
X(3872) ccTheoretical Predictions• Electromagnetic transitions for charmonium:
– c2 (11D1) (nS) forbidden (M2)
– c1 (23P1) [J/(2S)] allowed (E1)
– Predictions for relative rate varies, but are of similar order
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
19
X(3872) ccTheoretical Predictions• Radiative decays of the D0D0* molecule:
– Decay to J/ is possible in vector meson dominance scenario
– (2S) proceeds via annihilation, highly disfavoured
• Caveat: Theory-dependent, few definite predictions exist
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
20
X(3872) ccTheoretical Predictions• Radiative decays may discriminate between c2, c1(2P), and D0D0*
“Perhaps the most robust diagnostic is the ’ decay mode...Clearly a measurement of the J/ and ’ decay modes of the X(3872) will provide compelling clues to its internal structure.”
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
21
Analysis Description
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
22
The BaBar Experiment
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
23
The BaBar Experiment
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
24
Analysis Description• Search for radiative decays X(3872) [J/(2S)]
• Refit event with B mass constrained
• Variables of interest: mmiss = |pe+e- – pB|, mX, mK*
• Select candidate with best unconstrained mB
• Optimize selection cuts on MC using FOM:
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
25
Event Selection• B Meson Variables
– B Vertex P(2)
– Reconstructed B mass
• Kinematic Variables
– mmiss, mX
• cc Masses
– Unconstrained J/, (2S) mass
• Event shape
– Isotropy of momenta
• Photon variables
– Energy and shower-related
• Kaon variables
– Mass selection
– Decay length and vertexing
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
26
Signal Extraction Strategy• Fit strategy:
1. Unbinned 1(2)-D ML fit to mmiss (and mK* if applicable)
2. Use sPlot technique to get projection of signal events in mX
3. Fit to mX to extract number of signal events
• sPlot useful for isolating signal-like events for “bump-hunting”
• Test on well-known B c1,2 (J/) K decays first, then unblind for X(3872)
• Determine PDFs from truth-matched MC
• Fix most parameters except for normalization and background in mX fit
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
27
PDF Definitions - Signal• Selected the following PDF shapes:
– mmiss: Crystal Ball ( ~ 5 MeV/c2)
– mX: Double Gaussian ( ~ 5, 10 MeV/c2)
– mK*: BW * Gaussian ( ~ 45, 10 MeV/c2)
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
28
PDF Definition – Background• Bkgd. PDFs: mmiss: ARGUS + peak, mX: Linear (+ peak), mK*: Linear (+ peak)
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
29
sPlot Procedure• sPlot technique uses information from discriminant variables (mmiss) to unfold the
distribution in uncorrelated control variable (mX) by species (signal, background)
1. Conduct ML fit using PDFs (F) in variable (y) to get N events of species (
2. For species (a), each event (i) is assigned weight:
where Va is the covariance matrix between two species (a, )
3. Produce a histogram of the control variable using the weights for species
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
30
Signal Extraction Example• Begin with mixture of events
• Perform fit to mmiss to get
Nsig and Nbkgd
• Generate sPlots based on
the yield and distributions
• Can examine signal or
background species
• Fit sPlot for signal to
get X(3872) yield
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
31
Fit Validation• Test fit with MC toys:
– Insert expected number of signal events from MC
– Generate background samples from distributions
– Repeat N times
• Truth-matched MC trials
– Measure of bias: PDFs determined on truth, “true” events should work
• Non-truth-matched MC trials
– Measure “fit efficiency” (ie: how well PDFs match data and extract results)
• Other trials
– X(3872) with null result, 8 times estimated BF ((2S)), varying number of c2 events, with and without NR backgrounds, crossfeed from other K and c modes, X(3872) J/ + background, etc.
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
32
Efficiency• Determine signal extraction
efficiency in two stages
• Cut/Reconstruction efficiency
– Number of events from MC
reconstructed and passing all
selection cuts
• Fit efficiency from MC toy trials
– Described on previous slide
• General comment:
– decreases as N tracks increases
– Fit decreases as N increases and E decreases
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
33
Results and Implications
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
34
B c1,2 K Results• Observe signals in
all decay modes
• Consistent with PDG,
and best measurement for
K+, KS and K*+ modes
• Unexpected B0 c2 K*0
• Analysis of Run 1-3 subset
consistent with previous
BaBar result
• Set limits on other modes
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
35
B c1,2 K Systematic Uncertainties• Primarily for B c1 K; statistics-dominated for B c2 K
• Systematic uncertainties on the B c1 K yield are small
• Generally dominated by effects beyond this analysis
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
36
Summary of c1,2 K Results
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
37
X(3872) J/ Results• Find ~3.6 evidence for
X(3872) J/
• No evidence in other
decay modes
• Measure:
BF(B+ X(3872) K+) x
(X(3872) J/) =
(2.8+/-0.8+/-0.2) x 10-6
• Consistent with previous:
(3.3+/-1.0+/-0.3) x 10-6
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
38
X(3872) (2S) Results• First evidence for
X(3872) (2S)(~3.5 significance)
• Measure:
BF(B+ X(3872) K+) x
(X(3872) (2S) ) =
(9.5+/-2.7+/-0.9) x 10-6
• Ratio of BFs:
(X(3872) (2S) X(3872) J/
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
39
X(3872) Systematic Uncertainties• Entirely statistics-dominated in the X(3872) modes
• Most effects are related to the signal yield, not the BF calculation
• Largest concern is uncertainties in MC/data modeling
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
40
Summary of X(3872) Results
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
41
Full mX Range
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
42
Results and Implications• Results
– Best B c1K measurement, limits on c2, possible B c2 K*0 evidence
– Updated BF(B+ X(3872)K+)(X(3872) J/– First evidence for X(3872) (2S) – Accepted to PRL, to appear soon
• Hybrids and tetraquark already ruled out
• X(3872) (nS) disfavours 2-+ c2(1D) charmonium
• Large [(2S) J/ratio unexpected for D0D0*
• Consistent with 1++ c1(2P) charmonium
• Implications:
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
43
Conclusions?• Can X(3872) = c1(2P)fit into the charmonium model?
– c1(2P) mass prediction is too high
– Explanation for isospin violation and c1(2P) DD suppression
• Can X(3872) (2S) be reconciled with molecular model?
– Theoretical work needed to explain large BF(D0D*0 (nS) )
• What do we learn regarding X(3872) structure from this?
– Perhaps neither purely molecular nor charmonium
– D0D*0 – c1(2P) mixing or other phenomena?
The X(3872) saga continues...
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
44
Backup Slides
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
45
Charmonium Spectrum
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
46
X(3872) J/ Updates• Both Belle and BaBar have updated B X(3872)[XJ/] K results
• Belle finds no mass splitting, equal ratio between the B0 and B+ decay modes:
BaBar measurement confirms,
though suffers statistically
Sets limit on width: <3.3 MeV
2/)26.089.018.0( cMeVM X
05.022.082.0))3872((
))3872(( 00
KXBBR
KXBBRR
2/)4.06.17.2( cMeVM X
05.024.041.0))3872((
))3872(( 00
KXBBR
KXBBRR
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
47
X(3872) J/ Conflict• Belle searched for B X(3872)[XJ/ • 3-invariant mass shows enhancement assumed to be
• Separate BaBar analysis of B KJ/• Observe Y(3940), no evidence for X(3872)
• ie: X J/ not necessarily X J/
Radiative Decays of the X(3872) in BaBar, SLAC Experimental SeminarBryan Fulsom, March 3, 2009
48
X(3872) Null Results• Upper limits and null results:
– X J/ = consistent with “molecule”
– X DD = rules out 0+, 1-, 2+, ...
– X c1,2 rules out charmonium possibilities
– X J/= no charged partner