Search for Antineutrino Charged Current Coherent Pion Production at SciBooNE

Abstract

The SciBooNE experiment (Fermilab) recently published results of a search for charged current coherent pion production in neutrino mode: muon neutrinos scattering on carbon. The results of this study are that no evidence for coherent pion production is observed, and SciBooNE set 90% confidence level upper limits on the cross section ratio of charged current coherent pion production to the total charged current cross section. Recently proposed new coherent pion models predict a production of charged current coherent pion events just below the SciBooNE's upper limit. Motivated by this, we performed a search for charged current coherent pion production using SciBooNE's collected antineutrino data since antineutrino data are expected to be more sensitive to look at coherent pion production than neutrino data. This paper describes preliminary results of a search for antineutrino charged current coherent pion production at the SciBooNE experiment. Comment: 6 pages to be appeared in the proceedings for NuInt09 proceedings

Full text

Page 1

Search for Antineutrino Charged Current Coherent Pion
Production at SciBooNE
Hide-Kazu TANAKA
Massachusetts Institute of Technology, Department of Physics,
77 Massachusetts Avenue, Cambridge, MA 02139-4307
Abstract.
The SciBooNE experiment (Fermilab) recently published results of a search for charged current coherent pion production
in neutrino mode: muon neutrinos scattering on carbon, νµ12C → µ−12Cπ+. The results of this study are that no evidence
for coherent pion production is observed, and SciBooNE set 90% confidence level upper limits on the cross section ratio of
charged current coherent pion production to the total charged current cross section. Recently proposed new coherent pion
models predict a production of charged current coherent pion events just below the SciBooNE’s upper limit. Motivated by
this, we performed a search for charged current coherent pion production using SciBooNE’s collected antineutrino data since
antineutrino data are expected to be more sensitive to look at coherent pion production than neutrino data.
This paper describes preliminary results of a search for antineutrino charged current coherent pion production at the
SciBooNE experiment.
Keywords: neutrino-nucleus cross-section, coherent pion production
PACS: 01.30.Cc, 13.15.+g, 25.30.Pt
INTRODUCTION
The unknown neutrino mixing angle θ13is one of the most important goals in current neutrino experiments. For
the next generation of long baseline neutrino oscillation experiments, T2K and NOvA, the precise measurement of
neutrino-nucleus cross sections in the few GeV energy range is an essential ingredient in the interpretation of neutrino
oscillation signals. Charged current single charged-pion production (CC-1π) is a dominant background process for
νµ→ νXoscillation measurements.
Neutrino-induced charged current single charged pion production is dominated by baryonic resonance excitation off
a single nucleon bound in a nucleus in the neutrino energy region of a few GeV. The resonance state is followed by
its prompt decay into a nucleon and a pion in the final state. The process is written as νµN → µ−N?π+where N and
N?are proton or neutron. In addition to this reaction, neutrinos can produce pions by interacting coherently with the
nucleons forming the target nucleus. The process is expressed as νµA → µAπ+, where A is a nucleus. Understanding
the cross sections of these processes is important to study νµ→ νXoscillation (νµdisappearance) near one GeV.
Coherent pion production in neutrino-nucleus interactions has already been the subject of several experimental
campaigns. The neutrino energy range between 2 and 100 GeV has been investigated, including both the charged
current and neutral current modes, and using both neutrino and antineutrino probes [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]. The
results in the high energy region are well described by the Rein-Sehgal model [11] prediction that is widely used in
neutrino experiments today.
Recently, there are two new results on coherent pion production, and they are drawing a lot of attention in the
neutrino physics community. The non-existence of charged current coherent pion production in a 1.3 GeV wide-band
neutrino beam has been reported by K2K [12], while there exist charged current coherent pion production positive
results at higher neutrino energies. On the other hand, evidence for neutral current coherent pion production at a
similar neutrino energy has been reported by MiniBooNE [13].
The SciBooNE experiment recently published results [14] of a search for charged current coherent pion production
from muon neutrino scattering on carbon (based on the Rein-Sehgal model [11, 15]), with two distinct data samples;
averaged energy 1.1 GeV and 2.2 GeV. No evidence for coherent pion production was observed. SciBooNE set 90%
confidence level upper limits on the cross section ratio of charged current coherent pion production to the total charged
current cross section at 0.67×10−2at a mean neutrino energy of 1.1 GeV and 1.36×10−2at a mean neutrino energy
of 2.2 GeV.

Page 2

Along with the experimental studies of coherent pion production, new theoretical predictions for coherent pion
production also became available. Several theoretical models describing coherent pion production have been proposed,
e.g. [16, 17, 18, 19, 22, 21, 22], using different formalisms to describe the relevant physics. These new models predict
smaller charged current/neutral current coherent pion production cross sections than the Rein-Sehgal model prediction
and predict a charged current coherent pion cross section just below SciBooNE’s upper limit.
Because of the connection between neutrino and antineutrino coherent pion production processes in the theoretical
models, it will be interesting to repeat this analysis on SciBooNE’s already collected antineutrino data. Most models
predict a similar absolute cross section for neutrino and antineutrino coherent pion production, which means the ratio
of charged current coherent pion events to charged current inclusive events is expected to be larger in antineutrino data
because of the reduced total ¯ νcharged current event rate. Because of this, the antineutrino search has the potential to be
even more sensitive. Motivated by this, SciBooNE has performed a charged current coherent pion production search
in antineutrino mode.
SCIBOONE EXPERIMENT
The SciBooNE experiment [23] is designed to measure neutrino cross sections on carbon in the one GeV region. The
experiment collected data from June 2007 to August 2008 with neutrino and antineutrino beams in the FNAL Booster
Neutrino Beam line (BNB). The BNB uses a primary proton beam with kinetic energy 8 GeV and a beryllium target
placed inside an aluminum horn as a neutrino production target. The SciBooNE detector is located 100 m downstream
from the neutrino production target. The neutrino flux in the SciBooNE detector is dominated by muon neutrinos
(∼ 93%) in neutrino running mode. The flux-averaged mean neutrino energy is 0.7 GeV. When the horn polarity
is reversed, π−are focused and hence a predominantly antineutrino beam is created, with mean energy 0.6 GeV as
shown in Fig. 1.
TheSciBooNEdetectorconsistsofthreedetectorcomponents;SciBar,ElectromagneticCalorimeter(EC)andMuon
Range Detector (MRD). SciBar is a fully active and fine grained scintillator detector that consists of 14,336 bars
arranged in vertical and horizontal planes. SciBar is capable of detecting all charged particles and performing dE/dx-
based particle identification. The EC is located downstream of SciBar. The detector is a “spaghetti” calorimeter with
thickness of 11X0, and is used to measure π0and the intrinsic νecomponent of the neutrino beam. The MRD is located
downstream of the EC in order to measure the momentum of muons up to 1.2 GeV/c with range. It consists of 2-inch
thick iron plates sandwiched between layers of plastic scintillator planes.
SEARCH FOR ¯ ν CHARGED CURRENT COHERENT PION PRODUCTION
Antineutrino beam and wrong-sign background
Figure 1 shows the neutrino flux predictions [24] at the SciBooNE detector location in antineutrino mode running, as
a function of neutrino energy. As shown in the figure, a relatively large fraction of neutrino contamination is expected
in antineutrino mode running; ∼ 84% is muon-antineutrino, ∼ 16% muon-neutrino (cf. ∼ 93% pure muon-neutrino
beam in neutrino running mode). The beam simulation and the NEUT generator [25] predict about 30% contamination
from neutrinos in the total event rate (flux times cross section). Having precise knowledge of neutrino (“wrong-sign”)
backgrounds in data collected in antineutrino mode is important for any antineutrino cross section measurements.
Because SciBooNE does not have a magnetized detector, we cannot distinguish neutrino and antineutrino inter-
actions by the charge of the final state muons. However, event topology in antineutrino interactions differ from
neutrino’s. SciBar is uniquely suited to provide a measurement of the wrong-sign contamination in the antineutrino
beam. We exploit the fact that the fine-grained SciBar tracking detector can differentiate between final states with pro-
tons versus neutrons. Hence we can distinguish neutrino versus antineutrino charged current quasi elastic (CC-QE)
interactions on an event-by-event basis: νµ+n → µ−+ p and ¯ νµ+ p → µ++n. This is done based on their differing
final state composition: CC-QE neutrino interactions are expected to have two tracks (one each from the muon and
proton) while antineutrino interactions are expected to have only one track (from the muon), since the neutron is not
detected.
Figures 2 and 3 show the selected 1-track (µ) events and 2-track µ + p events in the antineutrino mode sample.
According to the MC simulation, the purity of right-sign (¯ ν) in the 1-track sample is ∼ 80%, and wrong-sign fraction

Page 3

10
-13
10
-12
10
-11
10
-10
00.51 1.522.5
Eν (GeV)
3
Flux (cm2 * pot * 25 MeV)-1
all
νe
ν
νµ
ν
–
e
–
µ
FIGURE 1.
is shown, as well as the flux per neutrino flavor: ¯ νµ(brown), νµ(green), νe(red), ¯ νe(blue). The averaged antineutrino energy is
predicted to be ∼0.6 GeV.
Antineutrino flux predictions at the SciBooNE detector, as a function of the neutrino energy. The total flux (black)
Entries 8966Entries 8966Entries 8966
(GeV/c)
µ
P
00 0.20.2 0.40.4 0.6 0.60.80.8 111.2 1.21.41.4
00
200
200
400
400
600
600
800
800
1000
1000
1200
1200
1400
1400
Entries 8966Entries 8966Entries 8966
(degree)
µ?
001010 2020 3030404050506060 707080809090
00
200
200
400
400
600600
800
800
1000
1000
1200
1200
1400
1400
1600
1600
CC QE
?
?
CC resonant
?
?
CC coherent
?
CC other
?
NC
?
(wrong sign)
?
BG (EC/MRD events)
FIGURE 2.
(µ) event. The right two plots show muon momentum (center) and muon angular (right) distributions of the sample. Data (with
statistical error) are shown as crosses.
The selected 1-track sample in the antineutrino mode sample. Left panel shows a typical event display of 1-track
in 2-track µ+p sample is ∼70%. Both samples shows fairly good agreement between data and MC simulation, which
means the flux prediction provides a reasonable understanding of wrong-sign contamination in antineutrino running
mode.
Entries Entries Entries 552 552 552
(GeV/c)
µ
P
00 0.20.20.4 0.40.60.60.80.8111.21.21.41.4
00
20
20
40
40
60
60
80
80
100
100
Entries Entries Entries 552 552 552
(degree)
µ?
001010202030 3040 4050 5060 6070708080 90 90
00
2020
40
40
60
60
80
80
100
100
CC QE
?
?
CC resonant
?
?
CC coherent
?
CC other
?
NC
?
(wrong sign)
?
BG (EC/MRD events)
FIGURE 3.
2-track µ + p event. The right two plots show muon momentum (center) and muon angular (right) distributions of the sample. Data
(with statistical error) are shown as crosses.
The selected 2-track µ + p sample in the antineutrino mode sample. The left panel shows a typical event display of

Page 4

Event selection
The event selection criteria used in this antineutrino charged current coherent pion search are identical to those used
in SciBooNE’s neutrino charged current coherent pion search [14].
The experimental signature of charged current single charged pion production is the existence of two and only two
tracks originating from a common vertex, both consistent with minimum ionizing particles (a muon and a charged
pion). Even in the case of a charged current resonant pion event with a proton in the final state, νµn → µ−pπ+, the
proton is often not reconstructed due to its low energy.
To identify charged current events, we search for tracks in SciBar matching with a track or hits in the MRD. Such
a track is defined as a SciBar-MRD matched track. The most energetic SciBar-MRD matched track in any event is
considered as the muon candidate. The matching criteria imposes a muon momentum threshold of 350 MeV/c. The
neutrino interaction vertex is reconstructed as the upstream edge of the muon candidate. We select events whose
vertices are in the SciBar fiducial volume (FV), 2.6 m×2.6 m×1.55 m, a total mass of 10.6 tons. Finally, event timing
is required to be within a 2 µsec beam timing window.
This analysis uses events with the muon stopping in the MRD, classified as MRD stopped events. The average
neutrino beam energy for true charged current events in the MRD stopped sample is ∼ 1.0 GeV.
Once the muon candidate and the neutrino interaction vertex are reconstructed, we search for other tracks origi-
nating from the vertex. Events with two and only two vertex-matched tracks are selected. The two track sample is
divided based on particle identification. The SciBar detector has the capability to distinguish protons from muons
and pions using dE/dx. The particle identification based on dE/dx is applied to both tracks to select events with two
MIP-like tracks (µ +π). The probability of mis-identification is estimated to be 1% for muons and 10% for protons.
The sample still contains CC-QE, mainly wrong-sign CC-QE, events in which a proton is mis-identified as a
minimum ionizing track. For each two-track event, we define an angle called ∆θpas the angle between the observed
second track direction and the expected proton track direction assuming a CC-QE event. Events with ∆θplarger than
20 degrees are selected to further reduce the CC-QE contamination. Further selections are applied in order to isolate
charged current coherent pion events from charged current resonant pion events which are the dominant backgrounds
for this analysis. In the case of charged current coherent pion events, both the muon and pion tracks are directed
forward. Events in which the track angle of the pion candidate with respect to the beam direction is less than 90
degrees are selected. To separate further charged current coherent pion events from charged current resonant pion
events, additional protons with momentum below the tracking threshold are detected by their large energy deposition
around the vertex, so-called vertex activity. We search for the maximum deposited energy in a strip around the vertex,
an area of 12.5 cm×12.5 cm in both views. Events with energy deposition greater than 10 MeV are considered to
have activity at the vertex. In this paper, we define “charged current coherent pion sample” which do not have activity
and its “counter-sample” that have activity.
Figure 4 and 5 show muon kinematics and reconstructed Q2distributions for the final selected charged current
coherent sample and its counter-sample, respectively. Here, the reconstructed Q2is calculated as
Q2
rec= 2Erec
ν(Eµ− pµcosθµ)−m2
µ
(1)
where Erec
ν
is the reconstructed neutrino energy calculated by assuming CC-QE kinematics,
Erec
ν
=1
2
(M2
p−m2
µ)−(Mn−V)2+2Eµ(Mn−V)
(Mn−V)−Eµ+ pµcosθµ
(2)
where Mpand Mnare the mass of proton and neutron, respectively, and V is the nuclear potential, which is set to
27 MeV. Note that in this analysis a MC tuning to constrain the background events is not done yet.
As shown in Fig. 4 there is a data deficit in the low Q2
agreement between data and MC simulation. The data clearly lie above the MC predicted backgrounds in the coherent
pion region below 0.1 (GeV/c)2, although the predicted coherent pion signal is larger than what is observed in the data.
These SciBooNE data suggest non-zero charged current coherent pion production, but it appears to be lower than that
predicted by the Rein-Sehgal model employed by the NEUT generator. The statistical significance of the data excess
is 4σ below 0.1 (GeV/c)2. It is interesting to note that the data excess above the predicted backgrounds are consistent
with the SciBooNE (and K2K) upper limits observed in the neutrino mode search within the statistical uncertainties.
Note that in this analysis the systematic uncertainties are not yet taken into account.
recregion, while the counter-sample, Fig 5, shows reasonable

Page 5

(GeV/c)
µ
P
000.20.20.4 0.4 0.60.6 0.80.8111.21.21.41.4
00
10
10
20
20
3030
4040
50
50
60
60
7070
(degree)
µ?
0010 1020 2030 3040 4050 5060 6070 7080 8090 90
00
10
10
20
20
30
30
40
40
50
50
60
60
70
70
2
(GeV/c)
rec.
2
Q
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
00
1010
2020
30
30
40
40
50
50
60
60
70
70
80
80
90
90
?
CC coherent
?
?
CC coherent
?
? (wrong sign)
CC other
?
?
CC resonant
?
CC QE
?
NC
?
BG (EC/MRD events)
FIGURE 4.
angular θµw.r.t. beam direction (center) and reconstructed Q2distributions of the sample.
Antineutrino charged current coherent pion sample. The left two plots show muon momentum (left) and muon
(GeV/c)
µ
P
000.20.20.40.40.60.60.80.8111.21.21.41.4
00
5
5
10
10
15
15
20
20
25
25
30
30
35
35
40
40
45
45
(degree)
µ?
0010 102020 30304040 50506060707080 809090
00
5
5
10
10
15
15
20
20
25
25
30
30
35
35
40
40
2
(GeV/c)
rec.
2
Q
000.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.911
00
10
10
20
20
30
30
40
40
50
50
60
60
70
70
80
80
?
CC coherent
?
?
CC coherent
?
? (wrong sign)
CC other
?
?
CC resonant
?
CC QE
?
NC
?
BG (EC/MRD events)
FIGURE 5.
and muon angular (center) and reconstructed Q2distributions of the sample.
The counter-sample to the charged current coherent pion sample (Fig. 4). The plots show muon momentum (left)
DISCUSSION
We have investigated the data excess further by examining other kinematic distributions. Figure 6 shows pion kine-
matics in the antineutrino charged current coherent pion sample (the same sample shown in Fig. 4). As shown in the
2-dimensional histogram in θµand θπ, where θµ/πare angles of muon and pion with respect to the beam direction,
the data excess concentrates in small θπand θµregion, in other words, the data excess is mainly visible at smaller
opening angles between muon and pion tracks.
Figure 7 shows the reconstructed event kinematics, Erec
ν
antineutrino charged current coherent pion sample for poin scattered angles less than 35 degrees and more than
35 degrees. As shown in Fig. 7, data in the sample θπ> 35◦are consistent with background prediction while there
and Q2
rec, distributions of two distinct subsamples of
Track angle (deg.)
002020 404060608080100 120 140 160 180100 120 140 160 180
00
10
10
2020
30
30
40
40
50
50
60
60
70
70
80
80
90
90
Opening angle (deg.)
00202040406060 8080 100 120 140 160 180100 120 140 160 180
00
10
10
20
20
30
30
40
40
50
50
60
60
70
70
-4
-3
-2
-1
0
1
2
3
(deg.)
µ
?
0 102030 40 506070 8090
(deg.)
?
?
0
10
20
30
40
50
60
70
80
90
FIGURE 6.
with respect to neutrino beam direction, plot in center shows opening angle between muon track and pion track. Right plot is a
2-dimensional (2D) histogram in θπvs. θµ. The 2D histogram shows the difference between data and MC simulation and divided
by statistical uncertainties. The MC simulation in the 2D histogram does not include charged current coherent pion prediction so
that the plot shows the data excess with respect to the background prediction of NEUT generator. For example, the color of orange
to red in the 2D histogram corresponds to 2- to 3-σ data excess.
Pion kinematics distributions in antineutrino charged current coherent pion sample. Left plot shows pion angle θπ