of 12
Measurements of direct
CP
asymmetries in
B
X
s
γ
decays using sum of
exclusive decays
J. P. Lees,
1
V. Poireau,
1
V. Tisserand,
1
E. Grauges,
2
A. Palano,
3a,3b
G. Eigen,
4
B. Stugu,
4
D. N. Brown,
5
L. T. Kerth,
5
Yu. G. Kolomensky,
5
M. J. Lee,
5
G. Lynch,
5
H. Koch,
6
T. Schroeder,
6
C. Hearty,
7
T. S. Mattison,
7
J. A. McKenna,
7
R. Y. So,
7
A. Khan,
8
V. E. Blinov,
9a,9c
A. R. Buzykaev,
9a
V. P. Druzhinin,
9a,9b
V. B. Golubev,
9a,9b
E. A. Kravchenko,
9a,9b
A. P. Onuchin,
9a,9c
S. I. Serednyakov,
9a,9b
Yu. I. Skovpen,
9a,9b
E. P. Solodov,
9a,9b
K. Yu. Todyshev,
9a,9b
A. N. Yushkov,
9a
D. Kirkby,
10
A. J. Lankford,
10
M. Mandelkern,
10
B. Dey,
11
J. W. Gary,
11
O. Long,
11
C. Campagnari,
12
M. Franco Sevilla,
12
T. M. Hong,
12
D. Kovalskyi,
12
J. D. Richman,
12
C. A. West,
12
A. M. Eisner,
13
W. S. Lockman,
13
B. A. Schumm,
13
A. Seiden,
13
D. S. Chao,
14
C. H. Cheng,
14
B. Echenard,
14
K. T. Flood,
14
D. G. Hitlin,
14
P. Ongmongkolkul,
14
F. C. Porter,
14
R. Andreassen,
15
Z. Huard,
15
B. T. Meadows,
15
B. G. Pushpawela,
15
M. D. Sokoloff,
15
L. Sun,
15
P. C. Bloom,
16
W. T. Ford,
16
A. Gaz,
16
U. Nauenberg,
16
J. G. Smith,
16
S. R. Wagner,
16
R. Ayad,
17
,
W. H. Toki,
17
B. Spaan,
18
R. Schwierz,
19
D. Bernard,
20
M. Verderi,
20
S. Playfer,
21
D. Bettoni,
22a
C. Bozzi,
22a
R. Calabrese,
22a,22b
G. Cibinetto,
22a,22b
E. Fioravanti,
22a,22b
I. Garzia,
22a,22b
E. Luppi,
22a,22b
L. Piemontese,
22a
V. Santoro,
22a
R. Baldini-Ferroli,
23
A. Calcaterra,
23
R. de Sangro,
23
G. Finocchiaro,
23
S. Martellotti,
23
P. Patteri,
23
I. M. Peruzzi,
23
,
M. Piccolo,
23
M. Rama,
23
A. Zallo,
23
R. Contri,
24a,24b
E. Guido,
24a,24b
M. Lo Vetere,
24a,24b
M. R. Monge,
24a,24b
S. Passaggio,
24a
C. Patrignani,
24a,24b
E. Robutti,
24a
B. Bhuyan,
25
V. Prasad,
25
M. Morii,
26
A. Adametz,
27
U. Uwer,
27
H. M. Lacker,
28
P. D. Dauncey,
29
U. Mallik,
30
C. Chen,
31
J. Cochran,
31
W. T. Meyer,
31
S. Prell,
31
H. Ahmed,
32
A. V. Gritsan,
33
N. Arnaud,
34
M. Davier,
34
D. Derkach,
34
G. Grosdidier,
34
F. Le Diberder,
34
A. M. Lutz,
34
B. Malaescu,
34
P. Roudeau,
34
A. Stocchi,
34
G. Wormser,
34
D. J. Lange,
35
D. M. Wright,
35
J. P. Coleman,
36
J. R. Fry,
36
E. Gabathuler,
36
D. E. Hutchcroft,
36
D. J. Payne,
36
C. Touramanis,
36
A. J. Bevan,
37
F. Di Lodovico,
37
R. Sacco,
37
G. Cowan,
38
J. Bougher,
39
D. N. Brown,
39
C. L. Davis,
39
A. G. Denig,
40
M. Fritsch,
40
W. Gradl,
40
K. Griessinger,
40
A. Hafner,
40
E. Prencipe,
40
K. R. Schubert,
40
R. J. Barlow,
41
G. D. Lafferty,
41
E. Behn,
42
R. Cenci,
42
B. Hamilton,
42
A. Jawahery,
42
D. A. Roberts,
42
R. Cowan,
43
D. Dujmic,
43
G. Sciolla,
43
R. Cheaib,
44
P. M. Patel,
44
,*
S. H. Robertson,
44
P. Biassoni,
45a,45b
N. Neri,
45a
F. Palombo,
45a,45b
L. Cremaldi,
46
R. Godang,
46
,**
P. Sonnek,
46
D. J. Summers,
46
M. Simard,
47
P. Taras,
47
G. De Nardo,
48a,48b
D. Monorchio,
48a,48b
G. Onorato,
48a,48b
C. Sciacca,
48a,48b
M. Martinelli,
49
G. Raven,
49
C. P. Jessop,
50
J. M. LoSecco,
50
K. Honscheid,
51
R. Kass,
51
J. Brau,
52
R. Frey,
52
N. B. Sinev,
52
D. Strom,
52
E. Torrence,
52
H. Ahmed,
32
E. Feltresi,
53a,53b
M. Margoni,
53a,53b
M. Morandin,
53a
M. Posocco,
53a
M. Rotondo,
53a
G. Simi,
53a
F. Simonetto,
53a,53b
R. Stroili,
53a,53b
S. Akar,
54
E. Ben-Haim,
54
M. Bomben,
54
G. R. Bonneaud,
54
H. Briand,
54
G. Calderini,
54
J. Chauveau,
54
Ph. Leruste,
54
G. Marchiori,
54
J. Ocariz,
54
S. Sitt,
54
M. Biasini,
55a,55b
E. Manoni,
55a
S. Pacetti,
55a,55b
A. Rossi,
55a
C. Angelini,
56a,56b
G. Batignani,
56a,56b
S. Bettarini,
56a,56b
M. Carpinelli,
56a,56b
,
††
G. Casarosa,
56a,56b
A. Cervelli,
56a,56b
F. Forti,
56a,56b
M. A. Giorgi,
56a,56b
A. Lusiani,
56a,56c
B. Oberhof,
56a,56b
E. Paoloni,
56a,56b
A. Perez,
56a
G. Rizzo,
56a,56b
J. J. Walsh,
56a
D. Lopes Pegna,
57
J. Olsen,
57
A. J. S. Smith,
57
R. Faccini,
58a,58b
F. Ferrarotto,
58a
F. Ferroni,
58a,58b
M. Gaspero,
58a,58b
L. Li Gioi,
58a
G. Piredda,
58a
C. Bünger,
59
O. Grünberg,
59
T. Hartmann,
59
T. Leddig,
59
C. Voß,
59
R. Waldi,
59
T. Adye,
60
E. O. Olaiya,
60
F. F. Wilson,
60
S. Emery,
61
G. Hamel de Monchenault,
61
G. Vasseur,
61
Ch. Yèche,
61
F. Anulli,
62
,
‡‡
D. Aston,
62
D. J. Bard,
62
J. F. Benitez,
62
C. Cartaro,
62
M. R. Convery,
62
J. Dorfan,
62
G. P. Dubois-Felsmann,
62
W. Dunwoodie,
62
M. Ebert,
62
R. C. Field,
62
B. G. Fulsom,
62
A. M. Gabareen,
62
M. T. Graham,
62
C. Hast,
62
W. R. Innes,
62
P. Kim,
62
M. L. Kocian,
62
D. W. G. S. Leith,
62
P. Lewis,
62
D. Lindemann,
62
B. Lindquist,
62
S. Luitz,
62
V. Luth,
62
H. L. Lynch,
62
D. B. MacFarlane,
62
D. R. Muller,
62
H. Neal,
62
S. Nelson,
62
M. Perl,
62
T. Pulliam,
62
B. N. Ratcliff,
62
A. Roodman,
62
A. A. Salnikov,
62
R. H. Schindler,
62
A. Snyder,
62
D. Su,
62
M. K. Sullivan,
62
J. Va
vra,
62
A. P. Wagner,
62
W. F. Wang,
62
W. J. Wisniewski,
62
M. Wittgen,
62
D. H. Wright,
62
H. W. Wulsin,
62
V. Ziegler,
62
W. Park,
63
M. V. Purohit,
63
R. M. White,
63
,§§
J. R. Wilson,
63
A. Randle-Conde,
64
S. J. Sekula,
64
M. Bellis,
65
P. R. Burchat,
65
T. S. Miyashita,
65
E. M. T. Puccio,
65
M. S. Alam,
66
J. A. Ernst,
66
R. Gorodeisky,
67
N. Guttman,
67
D. R. Peimer,
67
A. Soffer,
67
S. M. Spanier,
68
J. L. Ritchie,
69
A. M. Ruland,
69
R. F. Schwitters,
69
B. C. Wray,
69
J. M. Izen,
70
X. C. Lou,
70
F. Bianchi,
71a,71b
F. De Mori,
71a,71b
A. Filippi,
71a
D. Gamba,
71a,71b
S. Zambito,
71a,71b
L. Lanceri,
72a,72b
L. Vitale,
72a,72b
F. Martinez-Vidal,
73
A. Oyanguren,
73
P. Villanueva-Perez,
73
J. Albert,
74
Sw. Banerjee,
74
F. U. Bernlochner,
74
H. H. F. Choi,
74
G. J. King,
74
R. Kowalewski,
74
M. J. Lewczuk,
74
T. Lueck,
74
I. M. Nugent,
74
J. M. Roney,
74
R. J. Sobie,
74
N. Tasneem,
74
T. J. Gershon,
75
P. F. Harrison,
75
T. E. Latham,
75
H. R. Band,
76
S. Dasu,
76
Y. Pan,
76
R. Prepost,
76
and S. L. Wu
76
(
B
A
B
AR
Collaboration)
1
Laboratoire d
Annecy-le-Vieux de Physique des Particules (LAPP), Université de Savoie,
CNRS/IN2P3, F-74941 Annecy-Le-Vieux, France
2
Universitat de Barcelona, Facultat de Fisica, Departament ECM, E-08028 Barcelona, Spain
3a
INFN Sezione di Bari, I-70126 Bari, Italy
3b
Dipartimento di Fisica, Università di Bari, I-70126 Bari, Italy
PHYSICAL REVIEW D
90,
092001 (2014)
1550-7998
=
2014
=
90(9)
=
092001(12)
092001-1
© 2014 American Physical Society
4
University of Bergen, Institute of Physics, N-5007 Bergen, Norway
5
Lawrence Berkeley National Laboratory and University of California, Berkeley, California 94720, USA
6
Ruhr Universität Bochum, Institut für Experimentalphysik 1, D-44780 Bochum, Germany
7
University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z1
8
Brunel University, Uxbridge, Middlesex UB8 3PH, United Kingdom
9a
Budker Institute of Nuclear Physics SB RAS, Novosibirsk 630090, Russia
9b
Novosibirsk State University, Novosibirsk 630090, Russia
9c
Novosibirsk State Technical University, Novosibirsk 630092, Russia
10
University of California at Irvine, Irvine, California 92697, USA
11
University of California at Riverside, Riverside, California 92521, USA
12
University of California at Santa Barbara, Santa Barbara, California 93106, USA
13
University of California at Santa Cruz, Institute for Particle Physics, Santa Cruz, California 95064, USA
14
California Institute of Technology, Pasadena, California 91125, USA
15
University of Cincinnati, Cincinnati, Ohio 45221, USA
16
University of Colorado, Boulder, Colorado 80309, USA
17
Colorado State University, Fort Collins, Colorado 80523, USA
18
Technische Universität Dortmund, Fakultät Physik, D-44221 Dortmund, Germany
19
Technische Universität Dresden, Institut für Kern- und Teilchenphysik, D-01062 Dresden, Germany
20
Laboratoire Leprince-Ringuet, Ecole Polytechnique, CNRS/IN2P3, F-91128 Palaiseau, France
21
University of Edinburgh, Edinburgh EH9 3JZ, United Kingdom
22a
INFN Sezione di Ferrara, I-44122 Ferrara, Italy
22b
Dipartimento di Fisica e Scienze della Terra, Università di Ferrara, I-44122 Ferrara, Italy
23
INFN Laboratori Nazionali di Frascati, I-00044 Frascati, Italy
24a
INFN Sezione di Genova, I-16146 Genova, Italy
24b
Dipartimento di Fisica, Università di Genova, I-16146 Genova, Italy
25
Indian Institute of Technology Guwahati, Guwahati, Assam, 781 039, India
26
Harvard University, Cambridge, Massachusetts 02138, USA
27
Universität Heidelberg, Physikalisches Institut, D-69120 Heidelberg, Germany
28
Humboldt-Universität zu Berlin, Institut für Physik, D-12489 Berlin, Germany
29
Imperial College London, London, SW7 2AZ, United Kingdom
30
University of Iowa, Iowa City, Iowa 52242, USA
31
Iowa State University, Ames, Iowa 50011-3160, USA
32
Jazan University, Jazan 22822, Kingdom of Saudi Arabia
33
Johns Hopkins University, Baltimore, Maryland 21218, USA
34
Laboratoire de l
Accélérateur Linéaire, IN2P3/CNRS et Université Paris-Sud 11,
Centre Scientifique d
Orsay, F-91898 Orsay Cedex, France
35
Lawrence Livermore National Laboratory, Livermore, California 94550, USA
36
University of Liverpool, Liverpool L69 7ZE, United Kingdom
37
Queen Mary, University of London, London, E1 4NS, United Kingdom
38
University of London, Royal Holloway and Bedford New College, Egham, Surrey TW20 0EX,
United Kingdom
39
University of Louisville, Louisville, Kentucky 40292, USA
40
Johannes Gutenberg-Universität Mainz, Institut für Kernphysik, D-55099 Mainz, Germany
41
University of Manchester, Manchester M13 9PL, United Kingdom
42
University of Maryland, College Park, Maryland 20742, USA
43
Massachusetts Institute of Technology, Laboratory for Nuclear Science, Cambridge,
Massachusetts 02139, USA
44
McGill University, Montréal, Québec, Canada H3A 2T8
45a
INFN Sezione di Milano, I-20133 Milano, Italy
45b
Dipartimento di Fisica, Università di Milano, I-20133 Milano, Italy
46
University of Mississippi, University, Mississippi 38677, USA
47
Université de Montréal, Physique des Particules, Montréal, Québec, Canada H3C 3J7
48a
INFN Sezione di Napoli, I-80126 Napoli, Italy
48b
Dipartimento di Scienze Fisiche, Università di Napoli Federico II, I-80126 Napoli, Italy
49
NIKHEF, National Institute for Nuclear Physics and High Energy Physics, NL-1009 DB Amsterdam,
The Netherlands
50
University of Notre Dame, Notre Dame, Indiana 46556, USA
51
Ohio State University, Columbus, Ohio 43210, USA
52
University of Oregon, Eugene, Oregon 97403, USA
53a
INFN Sezione di Padova, I-35131 Padova, Italy
J. P. LEES
et al.
PHYSICAL REVIEW D
90,
092001 (2014)
092001-2
53b
Dipartimento di Fisica, Università di Padova, I-35131 Padova, Italy
54
Laboratoire de Physique Nucléaire et de Hautes Energies, IN2P3/CNRS,
Université Pierre et Marie Curie-Paris6, Université Denis Diderot-Paris7, F-75252 Paris, France
55a
INFN Sezione di Perugia, I-06123 Perugia, Italy
55b
Dipartimento di Fisica, Università di Perugia, I-06123 Perugia, Italy
56a
INFN Sezione di Pisa, I-56127 Pisa, Italy
56b
Dipartimento di Fisica, Università di Pisa, I-56127 Pisa, Italy
56c
Scuola Normale Superiore di Pisa, I-56127 Pisa, Italy
57
Princeton University, Princeton, New Jersey 08544, USA
58a
INFN Sezione di Roma, I-00185 Roma, Italy
58b
Dipartimento di Fisica, Università di Roma La Sapienza, I-00185 Roma, Italy
59
Universität Rostock, D-18051 Rostock, Germany
60
Rutherford Appleton Laboratory, Chilton, Didcot, Oxon OX11 0QX, United Kingdom
61
CEA, Irfu, SPP, Centre de Saclay, F-91191 Gif-sur-Yvette, France
62
SLAC National Accelerator Laboratory, Stanford, California 94309, USA
63
University of South Carolina, Columbia, South Carolina 29208, USA
64
Southern Methodist University, Dallas, Texas 75275, USA
65
Stanford University, Stanford, California 94305-4060, USA
66
State University of New York, Albany, New York 12222, USA
67
Tel Aviv University, School of Physics and Astronomy, Tel Aviv, 69978, Israel
68
University of Tennessee, Knoxville, Tennessee 37996, USA
69
University of Texas at Austin, Austin, Texas 78712, USA
70
University of Texas at Dallas, Richardson, Texas 75083, USA
71a
INFN Sezione di Torino, I-10125 Torino, Italy
71b
Dipartimento di Fisica, Università di Torino, I-10125 Torino, Italy
72a
INFN Sezione di Trieste, I-34127 Trieste, Italy
72b
Dipartimento di Fisica, Università di Trieste, I-34127 Trieste, Italy
73
IFIC, Universitat de Valencia-CSIC, E-46071 Valencia, Spain
74
University of Victoria, Victoria, British Columbia, Canada V8W 3P6
75
Department of Physics, University of Warwick, Coventry CV4 7AL, United Kingdom
76
University of Wisconsin, Madison, Wisconsin 53706, USA
(Received 4 June 2014; published 5 November 2014)
We measure the direct
CP
violation asymmetry,
A
CP
,in
B
X
s
γ
and the isospin difference of the
asymmetry,
Δ
A
CP
, using
429
fb
1
of data collected at
Υ
ð
4
S
Þ
resonance with the
BABAR
detector at the
PEP-II asymmetric-energy
e
þ
e
storage rings operating at the SLAC National Accelerator Laboratory.
B
mesons are reconstructed from ten charged
B
final states and six neutral
B
final states. We find
A
CP
¼þð
1
.
7

1
.
9

1
.
0
Þ
%
, which is in agreement with the Standard Model prediction and provides an
improvement on the world average. Moreover, we report the first measurement of the difference between
A
CP
for charged and neutral decay modes,
Δ
A
CP
¼þð
5
.
0

3
.
9

1
.
5
Þ
%
. Using the value of
Δ
A
CP
,we
also provide 68% and 90% confidence intervals on the imaginary part of the ratio of the Wilson coefficients
corresponding to the chromomagnetic dipole and the electromagnetic dipole transitions.
DOI:
10.1103/PhysRevD.90.092001
PACS numbers: 13.20.-v, 13.25.Hw
I. INTRODUCTION
The flavor-changing neutral current decay
B
X
s
γ
,
where
X
s
represents any hadronic system with one unit
of strangeness, is highly suppressed in the standard model
(SM), as is the direct
CP
asymmetry,
A
CP
¼
Γ
̄
B
0
=B
X
s
γ
Γ
B
0
=B
þ
X
̄
s
γ
Γ
̄
B
0
=B
X
s
γ
þ
Γ
B
0
=B
þ
X
̄
s
γ
;
ð
1
Þ
due to the combination of CKM and GIM suppressions
[1]
.
New physics effects could enhance the asymmetry to a
level as large as 15%
[2
4]
. The current world average of
*
Deceased.
Now at the University of Tabuk, Tabuk 71491, Saudi Arabia.
Also with Università di Perugia, Dipartimento di Fisica,
Perugia, Italy.
§
Now at Laboratoire de Physique Nucláire et de Hautes
Energies, IN2P3/CNRS, Paris, France.
Now at the University of Huddersfield, Huddersfield HD1
3DH, United Kingdom.
Now at University of South Alabama, Mobile, Alabama
36688, USA.
**
Also with Università di Sassari, Sassari, Italy.
††
Also with INFN Sezione di Roma, Roma, Italy.
‡‡
Now at Universidad Técnica Federico Santa Maria, Valpa-
raiso, Chile 2390123.
MEASUREMENTS OF DIRECT
CP
ASYMMETRIES IN
...
PHYSICAL REVIEW D
90,
092001 (2014)
092001-3
A
CP
based on the results from
BABAR
[5]
, Belle
[6]
and
CLEO
[7]
is
ð
0
.
8

2
.
9
Þ
%
[8]
. The SM prediction for the
asymmetry was found in a recent study to be long distance
dominated
[9]
and to be in the range
0
.
6%
<A
SM
CP
<
2
.
8%
.
Benzke
et al.
[9]
predict a difference in direct
CP
asymmetry for charged and neutral
B
mesons,
Δ
A
X
s
γ
¼
A
B

X
s
γ
A
B
0
=
̄
B
0
X
s
γ
;
ð
2
Þ
which suggests a new test of the SM. The difference,
Δ
A
X
s
γ
, arises from an interference term in
A
CP
that depends
on the charge of the spectator quark. The magnitude of
Δ
A
X
s
γ
is proportional to Im
ð
C
8
g
=C
7
γ
Þ
where
C
7
γ
and
C
8
g
are Wilson coefficients corresponding to the electromag-
netic dipole and the chromomagnetic dipole transitions,
respectively. The two coefficients are real in the SM;
therefore,
Δ
A
X
s
γ
¼
0
. New physics contributions from
the enhancement of the
CP
-violating phase or of the
magnitude of the two Wilson coefficients
[1,10]
, or the
introduction of new operators
[11]
, could enhance
Δ
A
X
s
γ
to
be as large as 10%
[9]
. Unlike
C
7
γ
,
C
8
g
currently does not
have a strong experimental constraint
[12]
. Thus a meas-
urement of
Δ
A
X
s
γ
together with the existing constraints on
C
7
γ
can provide a constraint on
C
8
g
.
Experimental studies of
B
X
s
γ
are approached in one
of two ways. The inclusive approach relies entirely on
observation of the high-energy photon from these decays
without reconstruction of the hadronic system
X
s
.By
ignoring the
X
s
system, this approach is sensitive to the
full
b
s
γ
decay rate and is robust against final state
fragmentation effects. The semi-inclusive approach recon-
structs the
X
s
system in as many specific final state
configurations as practical. This approach provides addi-
tional information, but since not all
X
s
final states can be
reconstructed without excessive background, fragmenta-
tion model-dependence is introduced if semi-inclusive
measurements are extrapolated to the complete ensemble
of
B
X
s
γ
decays.
BABAR
has recently published results
on the
B
X
s
γ
branching fraction and photon spectrum
for both approaches
[13,14]
. The inclusive approach has
also been used to search for direct
CP
violation, but since
the inclusive method does not distinguish hadronic final
states, decays due to
b
d
γ
transitions are included.
We report herein a measurement of
A
CP
and the first
measurement of
Δ
A
X
s
γ
using the semi-inclusive approach
with the full
BABAR
data set. We reconstruct 38 exclusive
B
-decay modes, listed in Table
I
, but for use in this analysis
a subset of 16 modes (marked with an asterisk in Table
I
)is
chosen for which high statistical significance is achieved.
Also, for this analysis, modes must be flavor self-tagging
(i.e., the bottomness can be determined from the recon-
structed final state). The 16 modes include ten charged
B
and six neutral
B
decays. After all event selection criteria
are applied, the mass of the hadronic
X
s
system (
m
X
s
)in
this measurement covers the range of about 0.6 to
2
.
0
GeV
=c
2
. The upper edge of this range approximately
corresponds to a minimum photon energy in the
B
rest
frame of 2.3 GeV. For
B
X
s
γ
decays with
0
.
6
<m
X
s
<
2
.
0
GeV
=c
2
, the ten charged
B
modes used
account for about 52% of all
B
þ
X
s
γ
decays and the six
neutral modes account for about 34% of all neutral
B
0
X
s
γ
decays.
1
In this analysis it is assumed that
A
CP
and
Δ
A
X
s
γ
are independent of final state fragmentation. That is,
it is assumed that
A
CP
and
Δ
A
X
s
γ
are independent of the
specific
X
s
final states used for this analysis and indepen-
dent of the
m
X
s
distribution of the selected events.
II. ANALYSIS OVERVIEW
With data from the
BABAR
detector (Sec.
III
), we
reconstructed
B
candidates from various final states
(Sec.
IV
). We then trained two multivariate classifiers
(Sec.
V
): one to separate correctly reconstructed
B
decays
from misreconstructed events and the other to reject the
continuum background,
e
þ
e
q
̄
q
, where
q
¼
u; d; s; c
.
The output of the first classifier is used to select the best
B
candidate for each event. Then, the outputs from both
classifiers are used to reject backgrounds. We use the
remaining events to determine the asymmetries.
We use identical procedures to extract three asymme-
tries: the asymmetries of charged and neutral
B
mesons,
and of the combined sample, and the difference,
Δ
A
X
s
γ
. The
bottomness of the
B
meson is determined by the charge of
the kaon for
B
0
and
̄
B
0
, and by the total charge of the
reconstructed
B
meson for
B
þ
and
B
.
We can decompose
A
CP
into three components:
A
CP
¼
A
peak
A
det
þ
D;
ð
3
Þ
where
A
peak
is the fitted asymmetry of the events in the
peak of the
m
ES
distribution (Sec.
VI
),
A
det
is the detector
asymmetry due to the difference in
K
þ
and
K
efficiency
(Sec.
VII
), and
D
is the bias due to peaking background
contamination (Sec.
VIII
). In this analysis we establish
upper bounds on the magnitude of
D
, and then treat those
as systematic errors.
III. DETECTOR AND DATA
We use a data sample of
429
fb
1
[15]
collected at the
Υ
ð
4
S
Þ
resonance,
ffiffiffi
s
p
¼
10
.
58
GeV
=c
2
, with the
BABAR
detector at the PEP-II asymmetric-energy
B
factory at the
SLAC National Accelerator Laboratory. The data corre-
spond to
471
×
10
6
produced
B
̄
B
pairs.
The
BABAR
detector and its operation are described in
detail elsewhere
[16,17]
. The charges and momenta of
charged particles are measured by a five-layer double-sided
1
If we include
K
L
modes as if they have same branching
fraction as
K
S
modes, the final state coverage for
0
.
6
<m
X
s
<
2
.
0
GeV
=c
2
is 69% for charged
B
and 34% for neutral
B
.
J. P. LEES
et al.
PHYSICAL REVIEW D
90,
092001 (2014)
092001-4
silicon strip detector (SVT) and a 40-layer drift chamber
(DCH) operated in a 1.5 T solenoidal field. Charged
K=
π
separation is achieved using
dE=dx
information from the
trackers and by a detector of internally reflected Cherenkov
light (DIRC), which measures the angle of the Cherenkov
radiation cone. An electromagnetic calorimeter (EMC)
consisting of an array of CsI(Tl) crystals measures the
energy of photons and electrons.
We use a Monte Carlo (MC) simulation based on
E
VT
G
EN
[18]
to optimize the event selection criteria. We
model the background as
e
þ
e
q
̄
q
,
e
þ
e
τ
þ
τ
and
B
̄
B
. We generate signal
B
X
s
γ
with a uniform photon
spectrum and then weight signal MC events so that the
photon spectrum matches the kinematic-scheme model
[19]
with parameter values consistent with the previous
BABAR
B
X
s
γ
photon spectrum analysis (
m
b
¼
4
.
65
GeV
=c
2
and
μ
2
π
¼
0
.
20
GeV
2
)
[20]
. We use JETSET
[21]
as the
fragmentation model and GEANT4
[22]
to simulate the
detector response.
IV.
B
RECONSTRUCTION
We reconstructed
B
meson candidates from 38 final
states listed in Table
I
. The 16 modes marked with an
asterisk (*) in Table
I
are used in the
CP
measurement. The
other final states are either not flavor-specific final states or
are low in yield. We reconstruct the unused modes in order
to veto them after selecting the best candidate. In total, we
use ten charged
B
final states and 6 neutral
B
final states in
the
A
CP
measurement. These final states are the same as
those used in a previous
BABAR
analysis
[5]
.
Charged kaons and pions are selected from tracks
classified with an error-correcting output code algorithm
[17,23]
. The classification uses SVT, DIRC, DCH, and
EMC information. The kaon particle identification (PID)
algorithm has approximately 90% efficiency and a pion-as-
kaon misidentification rate of about 1%. Pion identification
is roughly 99% efficient with a 15% kaon-as-pion mis-
identification rate.
Neutral kaons are reconstructed from the decay
K
0
S
π
þ
π
. The invariant mass of the two oppositely
charged tracks is required to be between 489 and 507 MeV.
The flight distance of the
K
0
S
must be greater than 0.2 cm
from the interaction point. The flight significance (defined
as the flight distance divided by the uncertainty in the flight
distance) of the
K
0
S
must be greater than three.
K
0
L
and
K
0
S
π
0
π
0
decays are not reconstructed for this analysis.
The neutral
π
0
and
η
mesons are reconstructed from two
photons. We require each photon to have energy of at least
30 MeV for
π
0
and at least 50 MeV for
η
. The invariant
mass of the two photons must be in the range of [115,150]
MeV for
π
0
candidates and in the range of [470,620] MeV
for
η
candidates. Only
π
0
and
η
candidates with momentum
greater than
200
MeV
=c
are used. We do not reconstruct
η
π
þ
π
π
0
decays explicitly, but some are included in
final states that contain
π
þ
π
π
0
.
Each event is required to have at least one photon with
energy
1
.
6
<E

γ
<
3
.
0
GeV, where the asterisk denotes
variables measured in the
Υ
ð
4
S
Þ
center-of-mass (CM)
frame. These photons are used as the primary photon in
reconstructing
B
mesons. Such a photon must have a lateral
moment
2
less than 0.8 and the nearest EMC cluster must be
at least 15 cm away. The angle of the photon momentum
with respect to the beam axis must satisfy
0
.
74
<
cos
θ
<
0
.
93
.
We make some further preliminary requirements to
reduce the data before giving events to the multivariate
classifiers for final selection. The invariant mass of
X
s
(all
daughters of the
B
candidate excluding the primary photon)
must satisfy
0
.
6
<m
X
s
<
3
.
2
GeV
=c
2
. The
X
s
candidate is
then combined with the primary photon to form a
B
candidate, which is required to have an energy-substituted
mass
m
ES
¼
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s=
4
p

B
2
p
, where
p

B
is the momentum of
B
in the CM frame, greater than
5
.
24
GeV
=c
2
. We also
require the difference between half of the beam total energy
and the energy of the reconstructed
B
in the CM frame,
j
Δ
E
j¼j
E

beam
=
2
E

B
j
, to be less than 0.15 GeV. The
angle between the thrust axis of the rest of the event(ROE)
and the primary photon must satisfy
j
cos
θ

T
γ
j
<
0
.
85
.
TABLE I. The 38 final states we reconstruct in this analysis.
Charge conjugation is implied. The 16 final states used in the
CP
measurement are marked with an asterisk.
Number
Final state
Number
Final state
1*
B
þ
K
S
π
þ
γ
20
B
0
K
S
π
þ
π
π
þ
π
γ
2*
B
þ
K
þ
π
0
γ
21
B
0
K
þ
π
þ
π
π
π
0
γ
3*
B
0
K
þ
π
γ
22
B
0
K
S
π
þ
π
π
0
π
0
γ
4
B
0
K
S
π
0
γ
23*
B
þ
K
þ
ηγ
5*
B
þ
K
þ
π
þ
π
γ
24
B
0
K
S
ηγ
6*
B
þ
K
S
π
þ
π
0
γ
25
B
þ
K
S
ηπ
þ
γ
7*
B
þ
K
þ
π
0
π
0
γ
26
B
þ
K
þ
ηπ
0
γ
8
B
0
K
S
π
þ
π
γ
27*
B
0
K
þ
ηπ
γ
9*
B
0
K
þ
π
π
0
γ
28
B
0
K
S
ηπ
0
γ
10
B
0
K
S
π
0
π
0
γ
29
B
þ
K
þ
ηπ
þ
π
γ
11*
B
þ
K
S
π
þ
π
π
þ
γ
30
B
þ
K
S
ηπ
þ
π
0
γ
12*
B
þ
K
þ
π
þ
π
π
0
γ
31
B
0
K
S
ηπ
þ
π
γ
13*
B
þ
K
S
π
þ
π
0
π
0
γ
32
B
0
K
þ
ηπ
π
0
γ
14*
B
0
K
þ
π
þ
π
π
γ
33*
B
þ
K
þ
K
K
þ
γ
15
B
0
K
S
π
0
π
þ
π
γ
34
B
0
K
þ
K
K
S
γ
16*
B
0
K
þ
π
π
0
π
0
γ
35
B
þ
K
þ
K
K
S
π
þ
γ
17
B
þ
K
þ
π
þ
π
π
þ
π
γ
36
B
þ
K
þ
K
K
þ
π
0
γ
18
B
þ
K
S
π
þ
π
π
þ
π
0
γ
37*
B
0
K
þ
K
K
þ
π
γ
19
B
þ
K
þ
π
þ
π
π
0
π
0
γ
38
B
0
K
þ
K
K
S
π
0
γ
2
The lateral moment is the ratio for the sum of energies of all
but the two most energetic crystals in the cluster weighted by the
squares of distances to the cluster center and the sum of energies
of all crystals weighted by the square of distance to the cluster
center.
MEASUREMENTS OF DIRECT
CP
ASYMMETRIES IN
...
PHYSICAL REVIEW D
90,
092001 (2014)
092001-5
V. EVENT AND CANDIDATE SELECTION
There are three main sources of background. The
dominant source is continuum background,
e
þ
e
q
̄
q
.
These events are more jet-like than the
e
þ
e
Υ
ð
4
S
Þ
B
̄
B
. Thus, event shape variables provide discrimination.
The continuum
m
ES
distribution does not peak at the
B
meson mass. The second background source is
B
̄
B
decays
to final states other than
X
s
γ
; hereafter we refer to these
as generic
B
̄
B
decays. The third source is a cross-feed
background which comes from actual
B
X
s
γ
decays
in which we fail to reconstruct the
B
in the correct final
state. The
e
þ
e
τ
þ
τ
contribution is negligibly small.
We first place a preliminary selection on the ratio of
angular moments
3
[24]
,
L
12
=L
10
<
0
.
46
to reduce the
number of the continuum background events. This ratio
measures the jettiness of the event. Since the mass of the
B
meson is close to half the mass of the
Υ
ð
4
S
Þ
, the kinetic
energy that the
B
meson can haveis less than that available to
e
þ
e
light quark pairs. Therefore, the signal peaks at a
lower value of
L
12
=L
10
than does the continuum
background.
The
B
meson reconstruction typically yields multiple
B
candidates per event (
10
on the average). To select the
best candidate, we train a random forest classifier
[25]
based on
Δ
E=
σ
E
, where
σ
E
is the uncertainty on the
B
candidate energy, the thrust of the reconstructed
B
candi-
date,
4
π
0
momentum, the invariant mass of the
X
s
system,
and the zeroth and fifth Fox-Wolfram moments
[26]
. This
signal selecting classifier (SSC) is trained on a large MC
event sample to separate correctly reconstructed
B
X
s
γ
decays from misreconstructed ones. For each event, the
candidate with the maximum classifier output is chosen as
the best candidate. This is the main difference from a
previous
BABAR
analysis
[5]
which chose the event with
the smallest
j
Δ
E
j
as the best candidate. This method
increases the efficiency by a factor of approximately two
for the same misidentification rate. For example, selecting
events with
m
ES
>
5
.
27
GeV
=c
2
, at a fake rate of 25%, the
signal rate is 16% for the
j
Δ
E
j
selection and 33% for
the SSC.
It should be emphasized that the best candidate selection
procedure also selects final states in which the bottomness
of the
B
cannot be deduced from the final decay products
(flavor-ambiguous final states). After selecting the best
candidate, we keep only events in which the best candidate
is reconstructed with the final states marked with an
asterisk in Table
I
. This removes events which are
flavor-ambiguous final states from the
A
CP
measurement.
Furthermore, because of the way the SSC was trained to
discriminate against misreconstructed
B
candidates, SSC
also provides good discriminating power against the
generic
B
̄
B
background.
To further reduce the continuum background we build
another random forest classifier, the background rejecting
classifier (BRC), using the following variables:
(i)
π
0
score: the output from a random forest classifier
using the invariant mass of the primary photon with
all other photons in the event and the energy of the
other photons, which is trained to reject high-energy
photons that come from the
π
0
γγ
decays.
(ii) Momentum flow
5
in 10° increments about the
reconstructed
B
direction.
(iii) Zeroth-, first- and second-order angular moments
along the primary photon axis computed in the CM
frame of the ROE.
(iv) The ratio of the second and the zeroth angular
moments described above.
(v)
j
cos
θ

B
j
: the cosine of the angle between the
B
flight
direction and the beam axis in the CM frame.
(vi)
j
cos
θ

T
j
: the cosine of the angle between the thrust
axis of the
B
candidate and the thrust axis of the
ROE in the CM frame.
(vii)
j
cos
θ

T
γ
j
: the cosine of the angle between the
primary photon momentum and the thrust axis of
the ROE in the CM frame.
To obtain the best sensitivity, we simultaneously opti-
mize, using MC samples, the SSC and BRC selections in
four
X
s
mass ranges ([0.6
1.1], [1.1
2.0], [2.0
2.4], and
½
2
.
4
2
.
8

GeV
=c
2
), maximizing
S=
ffiffiffiffiffiffiffiffiffiffiffiffi
S
þ
B
p
, where
S
is the
number of expected signal events and
B
is the number of
expected background events with
m
ES
>
5
.
27
GeV
=c
2
.
The optimized selection values are the same for both
B
and
̄
B
.
VI. FITTED ASYMMETRY
For each
B
flavor, we describe the
m
ES
distribution
with a sum of an ARGUS distribution
[27]
6
and a two-piece
normal distribution (
G
)
7
:
3
The Legendre moment of momentum for a given axis.
4
Thrust
¼
max
A

P
N
i
¼
1
j
A
·
p
i
j
P
N
i
¼
1
ffiffiffiffiffiffiffiffiffiffiffiffiffi
p
i
·
p
i
p

where
A
is a unit vector and
p
i
are three momenta of the decay
particles of the
B
candidate.
5
The scalar sum of all momenta within a cone of a given
opening angle about a given axis.
6
ARGUS
ð
x
;
c;
χ
;p
Þ¼
2
p
χ
2
ð
p
þ
1
Þ
Γ
ð
p
þ
1
Þ
Γ
ð
p
þ
1
;
1
2
χ
2
Þ
×
x
c
2

1
x
2
c
2

p
exp

1
2
χ
2

1
x
2
c
2

7
G
ð
x
;
μ
;
σ
L
;
σ
R
Þ¼
N
×
8
<
:
exp
n
ð
x
μ
Þ
2
2
σ
2
L
o
if
x<
μ
exp
n
ð
x
μ
Þ
2
2
σ
2
R
o
if
x
μ
;
where
N
is the normalization.
J. P. LEES
et al.
PHYSICAL REVIEW D
90,
092001 (2014)
092001-6
PDF
b
ð
m
ES
Þ¼
T
cont
2
ð
1
þ
A
cont
Þ
ARGUS
ð
m
ES
;
c
b
;
χ
b
;p
b
Þ
þ
T
peak
2
ð
1
þ
A
peak
Þ
G
ð
m
ES
;
μ
b
;
σ
b
L
;
σ
b
R
Þ
;
ð
4
Þ
PDF
̄
b
ð
m
ES
Þ¼
T
cont
2
ð
1
A
cont
Þ
ARGUS
ð
m
ES
;
c
̄
b
;
χ
̄
b
;p
̄
b
Þ
þ
T
peak
2
ð
1
A
peak
Þ
G
ð
m
ES
;
μ
̄
b
;
σ
̄
b
L
;
σ
̄
b
R
Þð
5
Þ
where
T
cont
¼
n
b
cont
þ
n
̄
b
cont
;
ð
6
Þ
T
peak
¼
n
b
peak
þ
n
̄
b
peak
ð
7
Þ
are the total number of events of both flavors described by
the ARGUS distribution and the two-piece normal distri-
bution and
A
cont
¼
n
b
cont
n
̄
b
cont
n
b
cont
þ
n
̄
b
cont
;
ð
8
Þ
A
peak
¼
n
b
peak
n
̄
b
peak
n
b
peak
þ
n
̄
b
peak
ð
9
Þ
are the flavor asymmetries of events described by the
ARGUS distribution and the two-piece normal distribution,
respectively. The superscript
b
and
̄
b
indicate whether
the parameter belongs to the
b
quark containing
B
meson
FIG. 1. The
m
ES
distributions along with fitted probability density functions, for:
̄
B
0
and
B
sample (top left),
B
0
and
B
þ
sample (top
right),
B
sample (middle left),
B
þ
sample (middle right),
̄
B
0
sample (bottom left), and
B
0
sample (bottom right). Data are shown as
points with error bars. The ARGUS distribution component, two-piece normal distribution component and the total probability density
function are shown with dotted lines, dashed lines, and solid lines, respectively.
MEASUREMENTS OF DIRECT
CP
ASYMMETRIES IN
...
PHYSICAL REVIEW D
90,
092001 (2014)
092001-7
(
̄
B
0
and
B
) distribution or the
̄
b
quark containing
B
meson
distribution (
B
0
and
B
þ
), respectively. In particular,
n
b
peak
and
n
̄
b
peak
are the numbers of events in the peaking
(Gaussian) part of the distribution. Similarly,
n
b
cont
and
n
̄
b
cont
are the numbers of events in the continuum (ARGUS)
part of the distribution. The shape parameters for ARGUS
distributions are the curvatures (
χ
b
and
χ
̄
b
), the powers
(
p
b
and
p
̄
b
), and the endpoint energies (
c
b
and
c
̄
b
). The
shape parameters for two-piece normal distribution are the
peak locations (
μ
b
and
μ
̄
b
), the left-side widths (
σ
b
L
and
σ
̄
b
L
),
and the right-side widths (
σ
b
R
and
σ
̄
b
R
).
It should be noted that
A
peak
is related to
A
CP
defined in
Eq.
(1)
by the relation shown in Eq.
(3)
. To obtain
A
peak
,we
perform a simultaneous binned likelihood fit for both
B
flavors. The ARGUS endpoint energies
c
b
and
c
̄
b
are fixed
at
5
.
29
GeV
=c
2
. All other shape parameters for the
ARGUS distributions and the two-piece normal distribu-
tions are allowed to float separately. Fig.
1
shows the
m
ES
distributions, along with fitted shapes. Table
II
summarizes
the results for
A
peak
.
VII. DETECTOR ASYMMETRY
Part of the difference between
A
peak
and
A
CP
comes from
the difference in
K
þ
and
K
efficiencies. The
K
þ
PID
efficiency is slightly higher than the
K
PID efficiency; the
difference also varies with the track momentum. The cause
of this difference is the fact that the cross section for
K
-
hadron interactions is higher than that for
K
þ
-hadron
interactions. This translates to the
K
having a greater
probability of interacting before it reaches the DIRC,
thereby lowering the quality of the
K
Cherenkov cone
angle measurement, which affects the PID performance.
The first order correction to
A
CP
from
K
þ
=K
efficiency
differences is given by
A
det
¼
ν
b
ν
̄
b
ν
b
þ
ν
̄
b
;
ð
10
Þ
where
ν
b
and
ν
̄
b
are the number of events for each flavor
after all selections, assuming the underlying physics has no
flavor asymmetry.
We use a sideband region (
m
ES
<
5
.
27
GeV
=c
2
) which
consists mostly of
e
þ
e
q
̄
q
events to measure
A
det
.We
do not expect a flavor asymmetry in the underlying physics
in this region. We count the number of events in the
sideband region for each flavor and use Eq.
(10)
to
determine
A
sideband
det
.
However, since the difference in
K
and
K
þ
hadron
cross section depends on
K
momentum and the
K
momentum distributions of the side band region and the
peaking region (
m
ES
>
5
.
27
GeV
=c
2
) slightly differ,
A
sideband
det
and
A
det
need not be identical. The variation of
A
det
for any
K
momentum distribution can be bounded by
the maximum and minimum value of the ratio between
K
þ
and
K
efficiencies (
ε
K
þ
=
ε
K
) in the
K
momentum range
of interest:
1
2

min
p
K
ε
K
þ
ε
K
1

A
det
1
2

max
p
K
ε
K
þ
ε
K
1

:
ð
11
Þ
The final states with no charged
K
can be considered as
having a special value of
p
K
where
ε
K
þ
and
ε
K
are
identical.
We use highly pure samples of charged kaons from the
decay
D
D
0
π
þ
, followed by
D
0
K
π
þ
, and its
charge conjugate, to measure the ratio of efficiencies for
K
þ
and
K
. We find that the deviation from unity of
ε
K
þ
=
ε
K
varies from 0% to 2.5% depending on the track
momentum.
The bound given in Eq.
(11)
implies that the distribution
of the differences between any two detector asymmetries
chosen uniformly within the bound is a triangle distribution
with the base width of 2.5%.
The standard deviation of such a distribution is
2
.
5%
=
ffiffiffiffiffi
24
p
¼
0
.
5%
. We use
A
sideband
det
as the central value
for
A
det
and this standard deviation as the systematic
uncertainty associated with detector asymmetry. Table
II
lists the results of
A
det
.
VIII. PEAKING BACKGROUND
CONTAMINATION
Our fitting procedure does not explicitly separate the
generic
B
̄
B
backgrounds and cross feed from the signal.
Both backgrounds have small peaking components, as
shown in Fig.
2
, so the yield for each flavor used in
calculating
A
peak
contains both signal and these peaking
backgrounds. We quantify the effect and include it as a
source of systematic uncertainty.
Let the number of signal events for
b
quark containing
B
mesons and
̄
b
quark containing
B
mesons be
n
b
and
n
̄
b
and
TABLE II. Summary of
A
CP
results along with
A
det
and systematic uncertainties due to peaking background contamination (D) for
each
B
sample. The
A
CP
s in the last column are calculated using Eq.
(3)
. The first error is statistical, the second (if present) is
systematics.
B
Sample
A
peak
D
A
det
A
CP
All
B
þð
0
.
33

1
.
87
Þ
%

0
.
88%
ð
1
.
40

0
.
49

0
.
51
Þ
%
þð
1
.
73

1
.
93

1
.
02
Þ
%
Charged
B
þð
3
.
14

2
.
86
Þ
%

0
.
80%
ð
1
.
09

0
.
67

0
.
51
Þ
%
þð
4
.
23

2
.
93

0
.
95
Þ
%
Neutral
B
ð
2
.
48

2
.
47
Þ
%

0
.
97%
ð
1
.
74

0
.
72

0
.
51
Þ
%
ð
0
.
74

2
.
57

1
.
10
Þ
%
J. P. LEES
et al.
PHYSICAL REVIEW D
90,
092001 (2014)
092001-8
the number of contaminating peaking background events
misreconstructed as
b
quark containing
B
mesons and
̄
b
quark containing
B
mesons be
β
b
and
β
̄
b
. The difference
between
A
peak
and
A
CP
due to peaking background con-
tamination is given by
D
¼
R
×
δ
A;
ð
12
Þ
where
R
is the ratio of the number of peaking background
events to the total number of events in the peaking region,
given by
R
¼
β
b
þ
β
̄
b
n
b
þ
n
̄
b
þ
β
b
þ
β
̄
b
;
ð
13
Þ
and
δ
A
is the difference between the true signal asymmetry
and the peaking background asymmetry, given by
δ
A
¼
n
b
n
̄
b
n
b
þ
n
̄
b
β
b
β
̄
b
β
b
þ
β
̄
b
:
ð
14
Þ
We estimate
R
using the MC sample. We use the sum of
the expected number of cross-feed background events and
expected number of generic
B
̄
B
events with
m
ES
>
5
.
27
GeV
=c
2
for each flavor as
β
b
and
β
̄
b
. We obtain
n
b
and
n
̄
b
from the total number of expected signal events for
each flavor.
Since the peaking background events are from misre-
constructed
B
mesons, the
m
ES
distribution of the peaking
background has a very long tail. It resembles the sum of an
ARGUS distribution and a small peaking part. The fit to the
total
m
ES
distribution is the sum of a two-piece normal
distribution and an ARGUS distribution. A significant
portion of peaking background is absorbed into the
FIG. 2. The contributions to the total
m
ES
distributions (gray lines with triangle markers) from the signal
B
X
s
γ
(gray lines with
circle markers), the continuum background (gray lines with x markers), the cross-feed background (gray lines with no marker), and the
generic
B
̄
B
background (solid black lines) according to the MC sample for:
̄
B
0
and
B
sample (top left),
B
0
and
B
þ
sample (top right),
B
sample (middle left),
B
þ
sample (middle right),
̄
B
0
sample (bottom left), and
B
0
sample (bottom right).
MEASUREMENTS OF DIRECT
CP
ASYMMETRIES IN
...
PHYSICAL REVIEW D
90,
092001 (2014)
092001-9