First observation of the decay D0→K−π+μ+μ−in the ρ0–ωregion of the dimuon mass spectrum

Physics Le e s B 757 (2016) 558–567
Con en s lis s a ailable a ScienceDi ec
Physics Le e s B
www.else ie .com/loca e/physle b
Fi s obse a ion o he decay D0→K−π+μ+μ−in he ρ0–ω egion
o he dimuon mass spec um
.LHCb Collabo a ion
a i c l e i n o a b s a c
A icle his o y:
Recei ed 29 Oc obe 2015
Recei ed in e ised o m 12 Ap il 2016
Accep ed 14 Ap il 2016
A ailable online 19 Ap il 2016
Edi o : L. Rolandi
A s udy o he decay D0→K−π+μ+μ−is pe o med using da a collec ed by he LHCb de ec o in
p o on–p o on collisions a a cen e-o -mass ene gy o 8TeV, co esponding o an in eg a ed luminosi y
o 2.0 b
−1. Decay candida es wi h muon pai s ha ha e an in a ian mass in he ange 675–875 MeV/c2
a e conside ed. This egion is domina ed by he ρ0and ω esonances. The b anching ac ion in his ange
is measu ed o be
B(D0→K−π+μ+μ−)=(4.17 ±0.12(s a )±0.40 (sys )) ×10−6.
This is he fi s obse a ion o he decay D0→K−π+μ+μ−. I s b anching ac ion is consis en wi h he
alue expec ed in he S anda d Model.
©2016 CERN o he benefi o he LHCb Collabo a ion. Published by Else ie B.V. This is an open access
a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/). Funded by SCOAP3.
1. In oduc ion
Ra e cha m decays may p oceed ia he highly supp essed
c→uμ+μ−fla ou changing neu al cu en p ocess. In he S an-
da d Model such p ocesses can only occu h ough loop diag ams,
whe e in cha m decays he GIM cancella ion [1] is almos com-
ple e. As a consequence, he sho -dis ance con ibu ion o he
inclusi e D →Xμ+μ−b anching ac ion is p edic ed o be as
low as O(10−9)[2], making hese decays in e es ing o sea ches
o new physics beyond he S anda d Model. Howe e , aking in o
accoun long-dis ance con ibu ions h ough ee diag ams in ol -
ing esonances such as D →XV(→μ+μ−), whe e V ep esen s a
φ, ρ0o ω ec o meson, he o al b anching ac ion o hese a e
cha m decays can each O(10−6)[2–4]. Thei sensi i i y o new
physics he e o e is g ea es in egions o he dimuon mass spec-
um away om hese esonances, whe e he main con ibu ions
o he b anching ac ion may come om sho -dis ance ampli-
udes. Angula asymme ies a e sensi i e o new physics bo h in
he icini y o hese esonances and away om hem [4–8] and
could be as la ge as O(1%).
This Le e ocuses on he measu emen o he decay1
D0→K−π+μ+μ−. This will p o ide an impo an e e ence
channel o measu emen s o he c→uμ+μ−p ocesses
D0→π+π−μ+μ−and D0→K+K−μ+μ−: p ecise b anching
ac ions a e easie o ob ain i hey a e compa ed wi h a no -
1The inclusion o cha ge conjuga e decays is implied.
malisa ion mode ha has simila ea u es. When es ic ed o he
dimuon mass ange 675 <m(μ+μ−) <875 MeV/c2, whe e he
ρ0and ω esonances a e expec ed o domina e, i can also be
used o no malise he decays D0→K−π+η()(→μ+μ−). Mea-
su ing hei b anching ac ions allows he coupling η()→μ+μ−
o be de e mined. This con ains c ucial in o ma ion o a ious
low ene gy phenomena, and is an inpu o he p edic ion o
he anomalous magne ic momen o he muon [9–11]. Focusing
on his dimuon mass ange also simplifies he analysis, which
does no ha e o accoun o he a ia ion o he selec ion effi-
ciency as a unc ion o m(μ+μ−). F om p e ious measu emen s
he mos s ingen 90% confidence le el uppe limi s on he de-
cay D0→K−π+μ+μ−a e se by he E791 expe imen [12]:
B(D0→K−π+μ+μ−) <35.9 ×10−5in he ull K−π+mass e-
gion and B(D0→K−π+μ+μ−) <2.4 ×10−5in he egion o he
K∗0 esonance.
The s udy p esen ed he e is based on da a collec ed by he
LHCb de ec o in p o on–p o on collisions a a cen e-o -mass
ene gy o 8TeV, co esponding o an in eg a ed luminosi y o
2.0 b
−1. Asubsample co esponding o an in eg a ed luminos-
i y o 1.6 b
−1has been used o measu e B(D0→K−π+μ+μ−).
The emainde o he da a se was used o op imise he selec ion.
The b anching ac ion B(D0→K−π+μ+μ−) is measu ed ela i e
o ha o he no malisa ion decay D0→K−π+π+π−. The mos
accu a e ecen measu emen o his b anching ac ion is used,
B(D0→K−π+π+π−) =(8.287 ±0.043 ±0.200) ×10−2, ob ained
by he CLEO expe imen [13].
h p://dx.doi.o g/10.1016/j.physle b.2016.04.029
0370-2693/©2016 CERN o he benefi o he LHCb Collabo a ion. Published by Else ie B.V. This is an open access a icle unde he CC BY license
(h p://c ea i ecommons.o g/licenses/by/4.0/). Funded by SCOAP3.
LHCb Collabo a ion / Physics Le e s B 757 (2016) 558–567 559
2. De ec o and simula ion
The LHCb de ec o [14,15] is a single-a m o wa d spec ome-
e co e ing he pseudo apidi y ange 2 <η<5, designed o he
s udy o pa icles con aining bo cqua ks. The de ec o includes
a high-p ecision acking sys em consis ing o a silicon-s ip e -
ex de ec o su ounding he pp in e ac ion egion, a la ge-a ea
silicon-s ip de ec o loca ed ups eam o a dipole magne wi h a
bending powe o abou 4Tm, and h ee s a ions o silicon-s ip
de ec o s and s aw d i ubes placed downs eam o he magne .
The acking sys em p o ides a measu emen o momen um, p, o
cha ged pa icles wi h a ela i e unce ain y ha a ies om 0.5%
a low momen um o 1.0% a 200 GeV/c. The minimum dis ance o
a ack o a p ima y e ex, he impac pa ame e (IP), is measu ed
wi h a esolu ion o (15 +29/pT)μm, whe e pTis he componen
o he momen um ans e se o he beam, in GeV/c.
Di e en ypes o cha ged had ons a e dis inguished using in-
o ma ion om wo ing-imaging Che enko de ec o s. Pho ons,
elec ons and had ons a e iden ified by a calo ime e sys em con-
sis ing o scin illa ing-pad and p eshowe de ec o s, an elec omag-
ne ic calo ime e and a had onic calo ime e . Muons a e iden ified
by a sys em composed o al e na ing laye s o i on and mul iwi e
p opo ional chambe s [16].
The online e en selec ion is pe o med by a igge [17],
which consis s o a ha dwa e s age, based on in o ma ion om
he calo ime e and muon sys ems, ollowed by a so wa e s age,
which applies a ull e en econs uc ion. In he offline selec ion,
equi emen s a e made on whe he he igge decision was due o
he signal candida e o o o he pa icles p oduced in he pp colli-
sion. Th oughou his Le e , hese wo non-exclusi e ca ego ies o
candida es a e e e ed o as T igge On Signal (TOS) and T igge
Independen o Signal (TIS) candida es.
Simula ed samples o D0→K−π+μ+μ−and
D0→K−π+π+π−decays ha e been p oduced. In he simula-
ion, pp collisions a e gene a ed using Py hia [18] wi h a spe-
cific LHCb configu a ion [19]. Decays o had onic pa icles a e
desc ibed by E Gen [20], in which final-s a e adia ion is gen-
e a ed using Pho os [21]. The in e ac ion o he gene a ed pa i-
cles wi h he de ec o , and i s esponse, a e implemen ed using
he Gean 4 oolki [22] as desc ibed in Re . [23]. No heo e i-
cal model o expe imen al measu emen p o ides a eliable de-
cay model o D0→K−π+μ+μ−. This decay mode is he e o e
modelled as an incohe en sum o esonan and non- esonan con-
ibu ions, such as K∗0→K−π+and ρ0/ω→μ+μ−, mo i a ed
by he esonan s uc u e obse ed in D0→K−π+π+π−and
D0→K−π+π+π−π0decays [24], and by he heo e ical p edic-
ions o Re . [4]. In he case o D0→K−π+π+π−, adecay model
ep oducing he da a was implemen ed using he MINT so wa e
package [25].
3. E en selec ion
The c i e ia used o selec he D0→K−π+μ+μ−and
D0→K−π+π+π−decays a e as simila as possible o allow
many sys ema ic unce ain ies o cancel in he efficiency a io.
A igge le el, only e en s ha a e TIS wi h espec o he
had on ha dwa e igge , which has a ans e se ene gy h eshold
o 3.7 GeV, a e kep . In he offline selec ion, he only di e ences
be ween he signal and no malisa ion channels a e he muon iden-
ifica ion c i e ia.
The fi s -le el so wa e igge selec s e en s ha con ain a
leas one good quali y ack wi h high pTand χ2
IP, whe e he la -
e is defined as he di e ence in χ2o he closes p ima y pp
in e ac ion e ex (PV) econs uc ed wi h and wi hou he pa i-
cle unde conside a ion. The offline selec ion equi es ha a leas
one o hese acks o igina es om ei he he D0→K−π+μ+μ−
o he D0→K−π+π+π−decay candida es. The second-le el
so wa e igge uses wo dedica ed selec ions o econs uc
D0→K−π+μ+μ−o D0→K−π+π+π−candida es o igina ing
om he PV. These combine good quali y acks ha
sa is y pT>350 MeV/cand p >3000 MeV/c. Amuon
(D0→K−π+μ+μ−) o cha ged had on (D0→K−π+π+π−) pai
is equi ed o o m a good quali y seconda y e ex ha is sig-
nifican ly displaced om he PV. In e en s whe e such a pai is
ound, wo cha ged had ons a e subsequen ly added. The esul -
ing ou -pa icle candida e mus ha e a good quali y e ex and i s
in a ian mass mus be consis en wi h he known D0mass [24].
The momen um ec o o his D0candida e mus be consis en
wi h ha ing o igina ed om he PV.
A p eselec ion ollows he igge selec ions. Fou cha ged pa i-
cles a e combined o o m D0candida es. T acks ha do no co e-
spond o ac ual ajec o ies o cha ged pa icles a e supp essed by
using a neu al ne wo k op imisa ion p ocedu e. To ejec he com-
bina o ial backg ound in ol ing acks om he PV, only high-p
and high-pT acks ha a e significan ly displaced om any PV
a e used. This backg ound is u he educed by equi ing ha
he ou decay p oduc s o he D0meson o m a good qual-
i y e ex ha is significan ly displaced om he PV and ha
pT(D0) >3000 MeV/c. These h ee c i e ia also ejec candida es
o med om pa ially econs uc ed cha m had on decays, com-
bined wi h ei he andom acks om he PV o wi h acks om
he decay o ano he cha med had on in he same e en . This
ype o backg ound is u he educed by equi ing he D0mo-
men um ec o is wi hin 14 m ad o he ec o ha joins he PV
wi h he D0decay e ex, ensu ing ha he D0candida e o ig-
ina es om he PV. Finally, he in a ian mass o he D0candi-
da e, which is econs uc ed wi h a esolu ion o abou 7MeV/c2,
is equi ed o lie wi hin 65 MeV/c2o he known D0mass. In
he case o D0→K−π+μ+μ−, m(μ+μ−)is es ic ed o he
ange 675–875 MeV/c2. The wo backg ounds desc ibed abo e a e
e e ed o as he non-peaking backg ound h oughou his Le -
e .
A e he p eselec ion, a mul i a ia e selec ion based on a
boos ed decision ee (BDT) [26,27] is used o u he supp ess he
non-peaking backg ound. The G adBoos algo i hm is used [28].
The BDT uses he ollowing a iables: he pTand χ2
IP o he final
s a e pa icles; he pTand χ2
IP o he D0candida e as well as he
χ2pe deg ee o eedom o i s e ex fi ; he significance o he
dis ance be ween his e ex and he PV; he la ges dis ance o
closes app oach be ween he acks ha o m he D0candida e;
he angle be ween he D0momen um ec o and he ec o ha
joins he PV wi h i s decay e ex. The cu on he BDT esponse
used in he selec ion disca ds mo e han 80% o he non-peaking
candida es and e ains mo e han 80% o he signal candida es ha
ha e passed he p eselec ion.
Finally, he in o ma ion om he RICH, he calo ime e s and he
muon sys ems a e combined o assign p obabili ies o each decay
p oduc o be a pion, a kaon o a muon, as desc ibed in Re . [15].
A loose equi emen on he kaon iden ifica ion p obabili y e-
jec s abou 90% o he backg ounds ha consis o π+π−μ+μ−
o π+π−π+π−combina ions while p ese ing 98% o he signal
candida es. In he case o D0→K−π+μ+μ−decays, he muon
iden ifica ion c i e ia ha e an efficiency o 90% pe signal muon
and educe he a e o misiden ified pions by a ac o o abou
150. In he absence o muon iden ifica ion, D0→K−π+π+π−
decays wi h wo misiden ified pions would ou numbe signal de-
cays by ou o de s o magni ude. A e hese pa icle iden ifica ion
equi emen s, his backg ound is educed o a ound 50% o he sig-
nal yield and is domina ed by decays in ol ing wo pion decays in
560 LHCb Collabo a ion / Physics Le e s B 757 (2016) 558–567
fligh (π+→μ+νμ). I is e e ed o as he peaking backg ound
h oughou his Le e .
In addi ion o D0→K−π+π+π−decays wi h wo misiden i-
fied pions, backg ounds due o he decays o D+, D+
s, D∗+, τ, Λ+
c
and 0
ca e conside ed. These a e s udied using simula ed e en s
and ound o be negligible.
The selec ion is op imised using da a and simula ed samples.
The BDT is ained using simula ed D0→K−π+μ+μ−e en s o
model he signal. The sample used o ep esen he backg ound
consis s o candida es wi h m(K+π−μ+μ−) >1890 MeV/c2,
d awn om 2% o he o al da a sample. Candida es on he low-
mass side o he signal peak a e no used due o he p es-
ence he e o peaking backg ound decays, whose ea u es a e
e y close o hose o signal decays. Op imal selec ion c i e-
ia on he BDT esponse and muon iden ifica ion a e ound
using ano he independen da a sample co esponding o 20%
o he o al da ase . The fi desc ibed in Sec . 4is used o
es ima e he yields o D0→K−π+μ+μ−signal (S), peaking
backg ound (Bpk) and non-peaking backg ound (Bnpk) p esen
in his sample in he egion o he signal peak, defined as
1840 <m(K−π+μ+μ−) <1890 MeV/c2. The equi emen s on he
muon iden ifica ion and BDT esponse a e chosen o maximise
S/S+Bpk +Bnpk.
The wo samples desc ibed abo e consis o e en s chosen
andomly om he 2012 da a and a e no used o he subse-
quen analysis. The emainde o he da ase (78%), which co e-
sponds o an in eg a ed luminosi y o 1.6 b
−1, is used o measu e
B(D0→K−π+μ+μ−). The final D0→K−π+μ+μ−sample ob-
ained wi h his selec ion consis s o 5411 candida es. In he case
o D0→K−π+π+π−, he la ge alue o B(D0→K−π+π+π−)
allows us o use a small sample (3 pb−1), d awn andomly om
he o al da ase . The final D0→K−π+π+π−sample consis s o
121 922 candida es.
4. De e mina ion o he D0→K−π+μ+μ−and
D0→K−π+π+π−yields
A simul aneous binned maximum likelihood fi o he
m(K−π+μ+μ−)and m(K−π+π+π−)dis ibu ions is pe o med
o measu e B(D0→K−π+μ+μ−).
In each sample, he p obabili y densi y unc ion (PDF) fi ed o
he signal peak is a Gaussian unc ion wi h powe law ails. I is
defined in he ollowing way:
(m;mD0,σ,αL,nL,αR,nR)
=
⎧
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎨
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎩
nL
|αL|nL×e−1
2α2
L
nL
|αL|−|αL|−m−mD0
σnL
i m−mD0
σ≤−|αL|,
nR
|αR|nR×e−1
2α2
R
nR
|αR|−|αR|+m−mD0
σnR
i m−mD0
σ≥|αR|,
exp −(m−mD0)2
2σ2o he wise,
whe e mD0and σa e he mean and wid h o he peak, and αL,
nL, αRand nRpa ame e ise he le and igh ails. This unc-
ion was ound o desc ibe accu a ely he m(K−π+μ+μ−)and
m(K−π+π+π−)dis ibu ions ob ained wi h he simula ion, which
exhibi non-Gaussian ails on bo h sides o he peaks. The ail on
he le -hand side is domina ed by final-s a e adia ion and in-
e ac ions wi h ma e , while he igh -hand side ail is due o
non-Gaussian e ec s in he econs uc ion.
The non-peaking backg ound in he D0→K−π+π+π−sam-
ple is desc ibed by a fi s -o de polynomial. In he case o
D0→K−π+μ+μ−, a second-o de polynomial is used.
Th ee peaking backg ounds due o misiden ified
D0→K−π+π+π−decays a e ca ego ised by he p esence o
candida es in ol ing misiden ified pions ha did no decay in
fligh be o e eaching he mos downs eam acking s a ions, o
candida es whe e one o wo pions decayed ups eam o hese
acking s a ions. Candida es om he fi s ca ego y a e desc ibed
by a one-dimensional ke nel densi y es ima e [29]. This PDF is
de i ed om he m(K−π+μ+μ−)dis ibu ion ob ained using
simula ed D0→K−π+π+π−decays econs uc ed unde he
D0→K−π+μ+μ−hypo hesis. Candida es om he emaining
wo ca ego ies appea as ails on he lowe -mass side o he
m(K−π+μ+μ−)dis ibu ion and mus be accoun ed o o a oid
biases in he non-peaking backg ound and in he signal yield mea-
su ed by he fi . Due o he small numbe o such candida es in he
simula ed sample, simula ed D0→K−π+π+π−candida es whe e
no pion decays in fligh a e al e ed o ep oduce he e ec o such
decays, and he co esponding m(K−π+μ+μ−)dis ibu ion is de-
e mined. This is achie ed by modi ying he momen um ec o s o
ei he one o wo o he pions p esen in he D0→K−π+π+π−
final s a e acco ding o he kinema ics o π+→μ+νμdecays. The
m(K−π+μ+μ−)dis ibu ions ob ained a e his modifica ion a e
con e ed in o one-dimensional ke nel densi y es ima es.
The fi model in ol es 5 yields: he signal yield, Nsig, he
yield o no malisa ion decays, ND0→K−π+π+π−, he peaking and
non-peaking backg ound yields, Npk and Nnpk, and he yield o
backg ound candida es in he D0→K−π+π+π−sample, NKπππ
npk .
They a e all ee pa ame e s in he fi . I also in ol es 15 pa ame-
e s o define he shapes o he PDFs. The pa ame e s desc ibing
he wid hs and uppe -mass ails a e ee pa ame e s in he fi
bu a e common be ween he PDFs o he D0→K−π+μ+μ−
and D0→K−π+π+π−peaks. The lowe -mass ail pa ame e s a e
de e mined sepa a ely. Those used o D0→K−π+π+π−can-
dida es a e allowed o a y in he fi . This is no possible o
D0→K−π+μ+μ−candida es because o he o e lap be ween
he signal and he D0→K−π+π+π−peaking backg ound and
he e o e he pa ame e s a e fixed o he alues ob ained om
he simula ed sample. In o al, he e a e 15 ee pa ame e s in he
fi .
The ela i e yields o he h ee peaking backg ound ca ego ies
desc ibed abo e a e fixed o alues ob ained by a fi o a la ge
con ol sample. I consis s o D0→K−π+μ+μ−candida es ha
a e in he TOS ca ego y wi h espec o he muon ha dwa e igge ,
in con as o he signal and no malisa ion samples ha a e in he
TIS ca ego y wi h espec o he had on igge . All o he o he
selec ion equi emen s a e he same as hose desc ibed in Sec . 3.
This TOS signal con ol sample consis s o 28 835 candida es and
con ains app oxima ely six imes mo e D0→K−π+μ+μ−decays
han he nominal TIS sample.
The fi esul s a e summa ised in Table 1 and he obse ed
mass dis ibu ions a e shown in Fig. 1, wi h fi p ojec ions o e -
laid. The main difficul ies in his p ocedu e a e he simila i-
ies in he shape o he signal, peaking backg ound and non-
peaking backg ound, and he o e lap be ween hei dis ibu ions
in m(K−π+μ+μ−). Howe e , hei impac on he measu emen
p esen ed in his Le e is limi ed, as can also be seen in Ta-
ble 1.
5. B anching ac ion measu emen
The b anching ac ion o he decay D0→K−π+μ+μ−is ob-
ained by combining he quan i ies p esen ed in Table 2 wi h he
b anching ac ion o he D0→K−π+π+π−decay acco ding o
LHCb Collabo a ion / Physics Le e s B 757 (2016) 558–567 561
Table 1
Summa y o he esul s o he fi desc ibed in Sec . 4. The yields measu ed
in he D0→K−π+μ+μ−sample and he co ela ions be ween hem, he
yields measu ed in he no malisa ion sample, he common wid h fi ed o he
D0→K−π+μ+μ−and D0→K−π+π+π−yields, and he ela i e unce ain y on
B(D0→K−π+μ+μ−) a e p esen ed. Unce ain ies on he fi ed pa ame e s a e
s a is ical. The a ia ion o he unce ain y on B(D0→K−π+μ+μ−)when he
backg ound yields a e fixed indica es o wha ex en i is enhanced by he need o
sepa a e con ibu ions in o e lap and which shapes p esen some simila i ies.
Pa ame e Value
Nsig 2357 ±67
Npk 1047 ±84
Nnpk 2007 ±116
ND0→K−π+π+π−83575 ±334
NKπππ
npk 38 346 ±257
σ7.17 ±0.03 MeV/c2
CNpk,Nnpk −78%
CNsig,Npk 27%
CNsig,Nnpk −48%
σB(D0→K−π+μ+μ−)2.9%
σB(D0→K−π+μ+μ−),i Npk fixed 2.8%
σB(D0→K−π+μ+μ−),i Npk and Nnpk fixed 2.4%
Fig. 1. Mass dis ibu ions o (a) D0→K−π+π+π−and (b) D0→K−π+μ+μ−
candida es. The da a a e shown as poin s (black) and he o al PDF (blue solid
line) is o e laid. In (a), he wo co esponding componen s o he fi model a e he
D0→K−π+π+π−decays ( ed solid line) and he non-peaking backg ound ( iole
dashed line). In (b), he componen s a e he D0→K−π+μ+μ−(long-dashed g een
line), he peaking backg ound due o misiden ified D0→K−π+π+π−decays ( ed
solid line), and he non-peaking backg ound ( iole dashed line). (Fo in e p e a ion
o he e e ences o colou in his figu e legend, he eade is e e ed o he web
e sion o his a icle.)
B(D0→K−π+μ+μ−)=ND0→K−π+μ+μ−
ND0→K−π+π+π−
×εD0→K−π+π+π−
εD0→K−π+μ+μ−
×B(D0→K−π+π+π−), (1)
Table 2
Measu ed efficiencies and yields o he decay D0→K−π+μ+μ−in he dimuon
mass ange 675–875 MeV/c2, and o he decay D0→K−π+π+π−. The unce -
ain ies a e s a is ical. In he case o efficiencies, i s ems om he fini e size o he
simula ed samples.
D0→K−π+μ+μ−D0→K−π+π+π−
Efficiency [10−5]8.8±0.28.2±0.1
Yields 2357 ±67 83 575 ±334
Table 3
Sys ema ic unce ain ies on B(D0→K−π+μ+μ−).
Sou ce Unce ain y [%]
T ack econs uc ion 3.2
Offline selec ion 2.0
Simula ed decay models 2.5
Ha dwa e igge 4.4
So wa e igge 4.3
Muon iden ifica ion 3.2
Kaon iden ifica ion 1.0
Size o simula ed sample 2.9
σsys (εD0→K−π+μ+μ−/εD0→K−π+π+π−)8.8
Signal shape pa ame e s 0.8
Peaking backg ound ails 1.5
Signal PDF 0.6
Non-peaking backg ound shape 2.1
σsys (NK−π+(μ+μ−)ρ0−ω/NK−π+π+π−)2.8
B(D0→K−π+π+π−)2.5
Quad a ic sum 9.6
whe e ND0→K−π+μ+μ−, ND0→K−π+π+π−, εD0→K−π+μ+μ−and
εD0→K−π+π+π−a e he yields and selec ion efficiencies o he sig-
nal and no malisa ion decays. The b anching ac ion o he signal
decay o dimuon in a ian masses in he ange 675–875 MeV/c2
is measu ed o be B(D0→K−π+μ+μ−) =(4.17 ±0.12)×10−6,
whe e he unce ain y is s a is ical.
5.1. Sys ema ic unce ain ies
The sys ema ic unce ain ies on B(D0→K−π+μ+μ−) a e
summa ised in Table 3. Those ela ed o econs uc ion and se-
lec ion efficiencies a e minimised hanks o he efficiency a io
in Eq. (1) and o he simila i ies be ween D0→K−π+μ+μ−
and D0→K−π+π+π−decays. This is illus a ed in Fig. 2,
which shows he dis ibu ions o he BDT esponse o he
D0→K−π+μ+μ−and D0→K−π+π+π−decays, bo h in
da a and simula ed samples. In da a, he backg ound con ibu-
ions a e emo ed using he sPlo echnique [30]. Also shown in
his figu e a e he a ios be ween he D0→K−π+μ+μ−and
D0→K−π+π+π−dis ibu ions. The BDT esponse, which com-
bines all he offline selec ion a iables (wi h he excep ion o
muon iden ifica ion c i e ia), is e y simila o bo h kinds o de-
cay and he di e ences a e well desc ibed by he simula ion. In
cases whe e selec ion c i e ia depend on he na u e o he decay
p oduc s, da a-d i en me hods a e used, as desc ibed below.
The unce ain y on he cha ged had on econs uc ion ineffi-
ciency is domina ed by he unce ain y on he p obabili y o un-
de go a nuclea in e ac ion in he de ec o . This inefficiency is
e alua ed using simula ed e en s. The co esponding unce ain y
is de i ed om he 10% unce ain y on he modelling o he de ec-
o ma e ial [31].
The selec ion efficiencies based on he kinema ical and geo-
me ical equi emen s a e de i ed om simula ion. A sys ema ic
unce ain y o ake in o accoun impe ec ack econs uc ion
562 LHCb Collabo a ion / Physics Le e s B 757 (2016) 558–567
Fig. 2. Dis ibu ions o he BDT esponse o D0→K−π+μ+μ−(ci cles) and
D0→K−π+π+π−decays ( iangles) in da a ( ull ma ke s) and simula ion (open
ma ke s). In da a, he backg ound con ibu ions a e emo ed using he sPlo
echnique. The lowe plo shows he a io be ween he D0→K−π+μ+μ−
and D0→K−π+π+π−dis ibu ions in da a ( ull squa es) and simula ion (open
squa es).
modelling is es ima ed by smea ing ack p ope ies o ep oduce
hose obse ed in da a. Simila ly, a sys ema ic unce ain y on he
efficiency o he BDT selec ion is assigned as he di e ence be-
ween he efficiency ob ained in da a and simula ion.
The unce ain ies in he decay models a e es ima ed sepa a ely
o he signal and no malisa ion channels. Fo he signal, his
is ca ied ou by eweigh ing simula ed D0→K−π+μ+μ−de-
cays o ep oduce he dis ibu ions o m(K−π+)and m(μ+μ−)
obse ed in da a, wi h he di e ence in efficiency ela i e o
he de aul being assigned as he sys ema ic unce ain y. Fo
D0→K−π+π+π−, he sensi i i y o he decay model is s ud-
ied by compa ing he de aul efficiency wi h ha ob ained in an
ex eme case in which he decay model p o ided by he MINT
package is eplaced by an incohe en sum o he esonances in-
ol ed in he decay, as gi en in Re . [24].
To a oid dependence on he modelling o he ha dwa e igge
in simula ion, i s efficiency is de e mined in da a. The efficiency o
be TIS wi h espec o had on ha dwa e igge is de e mined as
he ac ion o D0→K−π+μ+μ−decays ha ulfil his equi e-
men among D0→K−π+μ+μ−candida es ha a e TOS wi h e-
spec o he muon ha dwa e igge . I is measu ed in 12 di e en
egions defined in he (pT(D0), N ) plane, whe e N is he ack
mul iplici y o he e en . The o e all ha dwa e igge efficiency
o D0→K−π+μ+μ−decays is he a e age o hese 12 efficien-
cies weigh ed acco ding o he dis ibu ions o D0→K−π+μ+μ−
candida es obse ed in da a. The efficiency o he no malisa ion
mode is ob ained by weigh ing he same 12 efficiencies acco d-
ing o he dis ibu ions o D0→K−π+π+π−candida es. This
p ocedu e assumes ha he p obabili y o D0→K−π+μ+μ−
decays o ulfil he TIS equi emen is no enhanced by he e-
qui emen o also be in he TOS ca ego y and ha his TIS e -
ficiency is he same in e e y egion o D0→K−π+μ+μ−and
D0→K−π+π+π−decays. No di e ence is ound in simula ion
be ween he εD0→K−π+π+π−/εD0→K−π+μ+μ− a io ob ained wi h
his me hod and he a io o ue efficiencies, ob ained by di-
ec ly coun ing he numbe o simula ed D0→K−π+μ+μ−and
D0→K−π+π+π−decays ha ulfil he had on igge TIS e-
qui emen . To de e mine he sys ema ic unce ain y associa ed
wi h he ha dwa e igge efficiency, he unce ain y on his com-
pa ison is combined wi h he s a is ical unce ain ies on he 12
measu emen s pe o med in da a in (pT(D0), N ) egions.
A simila app oach is employed in he case o he fi s le el o
he so wa e igge . A sample o D0→K−π+π+π−candida es
is selec ed om da a ha sa isfied he igge equi emen s inde-
penden ly o hese candida es. The ac ion o D0→K−π+π+π−
decays whe e a leas one o he decay p oduc s also sa isfies he
equi emen s o his igge is measu ed using his sample. This e -
ficiency is measu ed in egions o pT(D0)and weigh ed acco ding
o dis ibu ions o his a iable in simula ed D0→K−π+μ+μ−
and D0→K−π+π+π−e en s. The a ia ion in he efficiency a-
io when hese dis ibu ions a e co ec ed o ma ch he da a is
used o e alua e he co esponding sys ema ic unce ain y.
The efficiency o he second-le el so wa e igge o he signal
decay is calcula ed ela i e o ha o he no malisa ion decay. This
a io is measu ed using D0→K−π+μ+μ−decays in da a and
simula ion and consis en esul s a e ob ained. The unce ain y on
his compa ison is he e o e assigned as he sys ema ic unce ain y
on his igge efficiency.
The efficiency o he muon iden ifica ion c i e ia is de e mined
in da a using a la ge and pu e sample o B →J/ψ(→μ+μ−)X
decays. Efficiencies measu ed in se e al egions o pT(μ), η(μ)
and N a e weigh ed acco ding o he dis ibu ion obse ed o
he muon candida es om D0→K−π+μ+μ−decays. Se e al de -
ini ions o hese domains a e conside ed, wi h a ying binnings.
The di e en efficiencies ob ained his way, as well as he effi-
ciencies ob ained in simula ed samples, a e compa ed o e alu-
a e he co esponding sys ema ic unce ain y. The same app oach
is used o e alua e he efficiency o he kaon iden ifica ion e-
qui emen . In his case, he calib a ion kaons a e p o ided by
D∗+ →D0(→K−π+)π+decays in da a.
In he fi ou lined in Sec . 4, he pa ame e s o he unc ion
ha desc ibe he lowe -mass ail o he D0→K−π+μ+μ−peak
a e fixed o alues ob ained om simula ion. The co esponding
sys ema ic unce ain y is de e mined by epea ing he fi using he
alues ob ained by a fi o he signal TOS con ol sample. A simila
di e ence is obse ed when he co esponding es is pe o med
o D0→K−π+π+π−candida es.
The sys ema ic unce ain y ela ed o he desc ip ion o he
peaking backg ound is de e mined by he change obse ed in
B(D0→K−π+μ+μ−) when he componen s due o he decay o
one o wo pions in fligh a e neglec ed, and when hei yields el-
a i e o he es o he peaking backg ound a e enhanced by wice
hei unce ain y.
Two o he sys ema ic unce ain ies ha e been e alua ed. To
es ima e he impac o he signal PDF employed, he fi is e-
pea ed using he C uij unc ion [32] ins ead. Po en ial e ec s
a ising om non-peaking backg ounds a e assessed by epea ing
he fi s wi h he non-peaking backg ounds assumed o be linea in
m(K−π+μ+μ−). The alues o he sys ema ic unce ain ies asso-
cia ed wi h he choice o fi model and i s pa ame e s we e also
u he alida ed using pseudoexpe imen s.
The impac on he fi o he simila i ies be ween he shapes
o he signal and backg ound componen s was u he con olled
in wo ways. Fi s , fixing he backg ound yields dec eases he el-
a i e unce ain y on B(D0→K−π+μ+μ−) om 2.9% o 2.4%.
This a ia ion is a lowe han he o al sys ema ic unce ain y
due o he yield de e mina ion (2.8%). Mo eo e , ano he s udy is
pe o med based on pseudoexpe imen s, gene a ed wi h ealis ic
alues o he yields and PDFs shape pa ame e s. The fi p o ed
able o e u n unbiased measu emen s o he gene a ed alue o
B(D0→K−π+μ+μ−) and an accu a e es ima ion o he s a is i-
cal unce ain y, consis en wi h he unce ain y ob ained in da a.

LHCb Collabo a ion / Physics Le e s B 757 (2016) 558–567 563
As can be seen in Table 3, he sys ema ic unce ain ies a e
domina ed by he unce ain y on he D0→K−π+μ+μ− o
D0→K−π+π+π−efficiency a io, which is la ge han he 2.9%
s a is ical unce ain y on B(D0→K−π+μ+μ−). As expec ed, his
sys ema ic unce ain y is p ima ily due o he di e en final s a e
pa icles o he wo decays. The igge efficiencies, and he muon
iden ifica ion and ack econs uc ion efficiencies, a e esponsi-
ble o abou 90% o his unce ain y. The unce ain ies due o he
yield de e mina ion and he knowledge o B(D0→K−π+π+π−)
ep esen seconda y con ibu ions.
6. Conclusions
The decay D0→K−π+μ+μ−is s udied using p o on–p o on
collision da a co esponding o an in eg a ed luminosi y o 2.0 b
−1
collec ed in 2012 by he LHCb de ec o a a cen e-o -mass ene gy
o 8TeV. The b anching ac ion o he decay D0→K−π+μ+μ−
in he dimuon mass ange 675–875 MeV/c2is measu ed o be
B(D0→K−π+μ+μ−)
=(4.17 ±0.12(s a ) ±0.40(sys ))×10−6.
This b anching ac ion can be compa ed o he S anda d Model
alue calcula ed in Re . [4], B(D0→K−π+μ+μ−) =6.7 ×10−6,
in he ull dimuon mass ange. This is he fi s obse a ion o
his decay. The b anching ac ion is measu ed wi h an o e all
p ecision o 10% and is one o de o magni ude lowe han he
p e ious mos s ingen uppe limi . P ecise measu emen s o he
D0→π+π−μ+μ−and D0→K+K−μ+μ−decays a e now pos-
sible in all egions o he dimuon in a ian mass since hey can be
compa ed wi h a no malisa ion mode ha has simila ea u es and
a p ecisely known b anching ac ion. This will allow mo e s in-
gen cons ain s on new physics o be ob ained using da a al eady
collec ed by he LHCb de ec o , and he sensi i i y o u u e expe -
imen s o angula asymme ies o be assessed.
The dis ibu ions o he K−π+and μ+μ−in a ian masses in
D0→K−π+μ+μ−decays a e shown in Fig. 3, whe e he back-
g ound con ibu ion is emo ed using he sPlo echnique [30],
aking he m(K−π+μ+μ−)in a ian mass as he disc imina ing
a iable. An ampli ude analysis would be equi ed o a ull un-
de s anding o he decay dynamics. The dis ibu ions in Fig. 3
sugges he p esence o addi ional con ibu ions, including he ω
esonance, beyond he K∗0ρ0in e media e s a e ha , acco ding o
Re . [4], should s ongly domina e he decay ampli ude.
Acknowledgemen s
We exp ess ou g a i ude o ou colleagues in he CERN ac-
cele a o depa men s o he excellen pe o mance o he LHC.
We hank he echnical and adminis a i e s a a he LHCb in-
s i u es. We acknowledge suppo om CERN and om he na-
ional agencies: CAPES, CNPq, FAPERJ and FINEP (B azil); NSFC
(China); CNRS/IN2P3 (F ance); BMBF, DFG and MPG (Ge many);
INFN (I aly); FOM and NWO (The Ne he lands); MNiSW and NCN
(Poland); MEN/IFA (Romania); MinES and FANO (Russia); MinECo
(Spain); SNSF and SER (Swi ze land); NASU (Uk aine); STFC (Uni ed
Kingdom); NSF (USA). We acknowledge he compu ing esou ces
ha a e p o ided by CERN, IN2P3 (F ance), KIT and DESY (Ge -
many), INFN (I aly), SURF (The Ne he lands), PIC (Spain), G idPP
(Uni ed Kingdom), RRCKI (Russia), CSCS (Swi ze land), IFIN-HH
(Romania), CBPF (B azil), PL-GRID (Poland) and OSC (USA). We
a e indeb ed o he communi ies behind he mul iple open sou ce
so wa e packages on which we depend. We a e also hank ul o
he compu ing esou ces and he access o so wa e R&D ools
p o ided by Yandex LLC (Russia). Indi idual g oups o membe s
Fig. 3. Backg ound sub ac ed dis ibu ion o (a) he K−π+in a ian mass and
(b) he μ+μ−in a ian mass, measu ed in D0→K−π+μ+μ−decays using he
sPlo echnique.
ha e ecei ed suppo om A H Founda ion (Ge many), EPLANET,
Ma ie Skłodowska-Cu ie Ac ions and ERC (Eu opean Union), Con-
seil Géné al de Hau e-Sa oie, Labex ENIGMASS and OCEVU, Ré-
gion Au e gne (F ance), RFBR (Russia), GVA, Xun aGal and GENCAT
(Spain), The Royal Socie y and Royal Commission o he Exhibi ion
o 1851 (Uni ed Kingdom).
Re e ences
[1] S.L. Glashow, J. Iliopoulos, L. Maiani, Weak in e ac ions wi h lep on–had on
symme y, Phys. Re . D 2 (1970) 1285.
[2] A. Paul, I.I. Bigi, S. Recksiegel, On D →Xul+l−wi hin he S anda d Model and
amewo ks like he li les Higgs model wi h T pa i y, Phys. Re . D 83 (2011)
114006, a Xi :1101.6053.
[3] S. Faj e , N. Ko˘
snik, S. P elo ˘
sek, Upda ed cons ain s on new physics in a e
cha m decays, Phys. Re . D 76 (2007) 074010, a Xi :0706.1133.
[4] L. Cappiello, O. Ca a, G. D’Amb osio, S anda d Model p edic ion and new
physics es s o D0→h+
1h−
2l+l−(h =π, K; l =e, μ), J. High Ene gy Phys. 04
(2013) 135, a Xi :1209.4235.
[5] A. Paul, A. De La Puen e, I.I. Bigi, Mani es a ions o wa ped ex a dimen-
sion in a e cha m decays and asymme ies, Phys. Re . D 90 (2014) 014035,
a Xi :1212.4849.
[6] S. Faj e , N. Ko˘
snik, P ospec s o disco e ing new physics in a e cha m decays,
a Xi :1510.00965.
564 LHCb Collabo a ion / Physics Le e s B 757 (2016) 558–567
[7] S. de Boe , G. Hille , Fla o & new physics oppo uni ies wi h a e cha m decays
in o lep ons, a Xi :1510.00311.
[8] S. Faj e , S. P elo ˘
sek, E ec s o li les Higgs model in a e D meson decays,
Phys. Re . D 73 (2006) 054026, a Xi :hep-ph/0511048.
[9] E. Kou, e al., No el app oach o measu e he lep onic η()→μ+μ−decays ia
cha med meson decays, h p://publica ion.lal.in2p3. /, 2016, LAL-16-011.
[10] P. Masjuan, P. Sanchez-Pue as, ηand ηdecays in o lep on pai s, a Xi :
1512.09292.
[11] A. Ny ele , On he p ecision o a da a-d i en es ima e o had onic ligh -
by-ligh sca e ing in he muon g-2: pseudoscala -pole con ibu ion, a Xi :
1602.03398.
[12] E791 Collabo a ion, E.M. Ai ala, e al., Sea ch o a e and o bidden cha m
meson decays D0→Vl+l−and hhll, Phys. Re . Le . 86 (2001) 3969, a Xi :hep-
ex/0011077.
[13] CLEO Collabo a ion, G. Bon icini, e al., Upda ed measu emen s o absolu e D+
and D0had onic b anching ac ions and σ(e+e−→DD)a Ecm =3774 MeV,
Phys. Re . D 89 (2014) 072002, a Xi :1312.6775.
[14] LHCb Collabo a ion, A.A. Al es J ., e al., The LHCb de ec o a he LHC, J. In-
s um. 3 (2008) S08005.
[15] LHCb Collabo a ion, R. Aaij, e al., LHCb de ec o pe o mance, In . J. Mod. Phys.
A 30 (2015) 1530022, a Xi :1412.6352.
[16] A.A. Al es J ., e al., Pe o mance o he LHCb muon sys em, J. Ins um. 8 (2013)
P02022, a Xi :1211.1346.
[17] R. Aaij, e al., The LHCb igge and i s pe o mance in 2011, J. Ins um. 8 (2013)
P04022, a Xi :1211.3055.
[18] T. Sjös and, S. M enna, P. Skands, PYTHIA 6.4 physics and manual, J. High En-
e gy Phys. 05 (2006) 026, a Xi :hep-ph/0603175;
T. Sjös and, S. M enna, P. Skands, A b ie in oduc ion o PYTHIA 8.1, Compu .
Phys. Commun. 178 (2008) 852, a Xi :0710.3820.
[19] I. Belyae , e al., Handling o he gene a ion o p ima y e en s in Gauss, he
LHCb simula ion amewo k, J. Phys. Con . Se . 331 (2011) 032047.
[20] D.J. Lange, The E Gen pa icle decay simula ion package, Nucl. Ins um. Me h-
ods A 462 (2001) 152.
[21] P. Golonka, Z. Was, PHOTOS Mon e Ca lo: ap ecision ool o QED co ec ions
in Zand Wdecays, Eu . Phys. J. C 45 (2006) 97, a Xi :hep-ph/0506026.
[22] Gean 4 Collabo a ion, J. Allison, e al., Gean 4 de elopmen s and applica ions,
IEEE T ans. Nucl. Sci. 53 (2006) 270;
Gean 4 Collabo a ion, S. Agos inelli, e al., Gean 4: a simula ion oolki , Nucl.
Ins um. Me hods A 506 (2003) 250.
[23] M. Clemencic, e al., The LHCb simula ion applica ion, Gauss: design, e olu ion
and expe ience, J. Phys. Con . Se . 331 (2011) 032023.
[24] Pa icle Da a G oup, K.A. Oli e, e al., Re iew o pa icle physics, Chin. Phys. C
38 (2014) 090001.
[25] P. Nason, MINT: a compu e p og am o adap i e Mon e Ca lo in eg a ion and
gene a ion o unweigh ed dis ibu ions, a Xi :0709.2085.
[26] L. B eiman, J.H. F iedman, R.A. Olshen, C.J. S one, Classifica ion and Reg ession
T ees, Wadswo h In e na ional G oup, Belmon , CA, USA, 1984.
[27] B.P. Roe, e al., Boos ed decision ees as an al e na i e o a ificial neu al ne -
wo ks o pa icle iden ifica ion, Nucl. Ins um. Me hods A 543 (2005) 577,
a Xi :physics/0408124.
[28] A. Hoecke , e al., TMVA – oolki o mul i a ia e da a analysis, PoS ACAT
(2007) 040, a Xi :physics/0703039.
[29] K.S. C anme , Ke nel es ima ion in high-ene gy physics, Compu . Phys. Com-
mun. 136 (2001) 198, a Xi :hep-ex/0011057.
[30] M. Pi k, F.R. Le Dibe de , sPlo : a s a is ical ool o un old da a dis ibu ions,
Nucl. Ins um. Me hods A 555 (2005) 356, a Xi :physics/0402083.
[31] LHCb Collabo a ion, R. Aaij, e al., Measu emen o he ack econs uc ion e -
ficiency a LHCb, J. Ins um. 10 (2015) P02007, a Xi :1408.1251.
[32] BaBa Collabo a ion, P. del Amo Sanchez, e al., S udy o B →Xγdecays and
de e mina ion o |V d/V s|, Phys. Re . D 82 (2010) 051101, a Xi :1005.4087.
LHCb Collabo a ion
R. Aaij 39, C. Abellán Be e a 41, B. Ade a 38, M. Adinolfi 47, A. A olde 53, Z. Ajal ouni 5, S. Aka 6,
J. Alb ech 10, F. Alessio 39, M. Alexande 52, S. Ali 42, G. Alkhazo 31, P. Al a ez Ca elle 54, A.A. Al es J 58,
S. Ama o 2, S. Ame io 23, Y. Amhis 7, L. An 3, L. Ande lini 18, J. Ande son 41, G. And eassi 40,
M. And eo i 17, , J.E. And ews 59, R.B. Appleby 55, O. Aquines Gu ie ez 11, F. A chilli 39, P. d’A gen 12,
A. A amono 36, M. A uso 60, E. Aslanides 6, G. Au iemma 26,m, M. Baalouch 5, S. Bachmann 12,
J.J. Back 49, A. Badalo 37, C. Baesso 61, W. Baldini 17,39, R.J. Ba low 55, C. Ba schel 39, S. Ba suk 7,
W. Ba e 39, V. Ba ozskaya 29, V. Ba is a 40, A. Bay 40, L. Beaucou 4, J. Beddow 52, F. Bedeschi 24,
I. Bediaga 1, L.J. Bel 42, V. Bellee 40, N. Belloli 21,j, I. Belyae 32, E. Ben-Haim 8, G. Benci enni 19,
S. Benson 39, J. Ben on 47, A. Be ezhnoy 33, R. Be ne 41, A. Be olin23, M.-O. Be le 39,
M. an Beuzekom 42, A. Bien12, S. Bi ani 46, P. Billoi 8, T. Bi d 55, A. Bi nk au 10, A. Bizze i 18,h,
T. Blake 49, F. Blanc 40, J. Blouw 11, S. Blusk 60, V. Bocci 26, A. Bonda 35,69, N. Bonda 31,39, W. Boni en o 16,
S. Bo ghi 55, M. Bo sa o 7, T.J.V. Bowcock 53, E. Bowen 41, C. Bozzi 17, S. B aun 12, M. B i sch 11,
T. B i on 60, J. B odzicka 55, N.H. B ook 47, E. Buchanan 47, C. Bu 55, A. Bu sche41, J. Buy ae 39,
S. Cadeddu 16, R. Calab ese 17, , M. Cal i 21,j, M. Cal o Gomez 37,o, P. Campana 19, D. Campo a Pe ez 39,
L. Cap io i 55, A. Ca bone15,d, G. Ca boni 25,k, R. Ca dinale 20,i, A. Ca dini 16, P. Ca ni i 21,j, L. Ca son 51,
K. Ca alho Akiba 2,39, G. Casse 53, L. Cassina 21,j, L. Cas illo Ga cia 40, M. Ca aneo 39, Ch. Caue 10,
G. Ca alle o 20, R. Cenci 24,s, M. Cha les 8, Ph. Cha pen ie 39, M. Che de ille 4, S. Chen 55, S.-F. Cheung 56,
N. Chiapolini 41, M. Ch zaszcz 41, X. Cid Vidal 39, G. Cieza ek 42, P.E.L. Cla ke 51, M. Clemencic 39,
H.V. Cli 48, J. Closie 39, V. Coco 39, J. Cogan 6, E. Cogne as 5, V. Cogoni 16,e, L. Cojoca iu 30, G. Collazuol 23,
P. Collins 39, A. Come ma-Mon ells 12, A. Con u16, A. Cook 47, M. Coombes 47, S. Coque eau 8, G. Co i 39,
M. Co o 17, , B. Cou u ie 39, G.A. Cowan 51, D.C. C aik 49, A. C ocombe 49, M. C uz To es 61, S. Cunli e 54,
R. Cu ie 54, C. D’Amb osio 39, E. Dall’Occo 42, J. Dalseno 47, P.N.Y. Da id 42, A. Da is 58,
O. De Aguia F ancisco 2, K. De B uyn 6, S. De Capua 55, M. De Cian 12, J.M. De Mi anda 1, L. De Paula 2,
P. De Simone 19, C.-T. Dean 52, D. Decamp 4, M. Deckenho 10, L. Del Buono 8, N. Déléage 4, M. Demme 10,
D. De kach 66, O. Deschamps 5, F. De o i 39, B. Dey 22, A. Di Can o39, F. Di Ruscio 25, H. Dijks a 39,
S. Donlea y 53, F. Do dei 12, M. Do igo 40, A. Dosil Suá ez 38, D. Dosse 49, A. Do bnya 44, K. D eimanis 53,
L. Du ou 42, G. Dujany 55, F. Dupe uis 40, P. Du an e 39, R. Dzhelyadin 36, A. Dziu da 27, A. Dzyuba31,
S. Easo 50,39, U. Egede 54, V. Ego yche 32, S. Eidelman 35,69, S. Eisenha d 51, U. Ei schbe ge 10,
R. Ekelho 10, L. Eklund 52, I. El Ri ai 5, Ch. Elsasse 41, S. Ely 60, S. Esen 12, H.M. E ans 48, T. E ans 56,
LHCb Collabo a ion / Physics Le e s B 757 (2016) 558–567 565
A. Falabella 15, C. Fä be 39, N. Fa ley 46, S. Fa y 53, R. Fay 53, D. Fe guson 51, V. Fe nandez Albo 38,
F. Fe a i 15, F. Fe ei a Rod igues 1, M. Fe o-Luzzi 39, S. Filippo 34, M. Fio e 17,39, , M. Fio ini 17, ,
M. Fi lej 28, C. Fi zpa ick 40, T. Fiu owski 28, K. Fohl 39, P. Fol 54, M. Fon ana 16, F. Fon anelli 20,i,
D.C. Fo shaw 60, R. Fo y 39, M. F ank 39, C. F ei 39, M. F osini 18, J. Fu 22, E. Fu a o 25,k,
A. Gallas To ei a 38, D. Galli 15,d, S. Gallo ini 23, S. Gambe a 51, M. Gandelman 2, P. Gandini 56, Y. Gao 3,
J. Ga cía Pa diñas 38, J. Ga a Tico 48, L. Ga ido 37, D. Gascon 37, C. Gaspa 39, R. Gauld 56, L. Ga a di 10,
G. Gazzoni 5, D. Ge ick 12, E. Ge sabeck 12, M. Ge sabeck 55, T. Ge shon 49, Ph. Ghez 4, S. Gianì 40,
V. Gibson 48, O.G. Gi a d 40, L. Giubega 30, V.V. Gligo o 39, C. Göbel 61, D. Golubko 32, A. Golu in 54,39,
A. Gomes1,a, C. Go i 21,j, M. G abalosa Gánda a 5, R. G aciani Diaz37, L.A. G anado Ca doso 39,
E. G augés 37, E. G a e ini 41, G. G aziani 18, A. G ecu30, E. G eening 56, S. G egson 48, P. G iffi h 46,
L. G illo 12, O. G ünbe g 64, B. Gui 60, E. Gushchin 34, Yu. Guz 36,39, T. Gys 39, T. Hada izadeh 56,
C. Hadji asiliou 60, G. Hae eli 40, C. Haen 39, S.C. Haines 48, S. Hall 54, B. Hamil on 59, X. Han 12,
S. Hansmann-Menzeme 12, N. Ha new 56, S.T. Ha new 47, J. Ha ison 55, J. He 39, T. Head 40, V. Heijne 42,
A. Heis e 9, K. Hennessy 53, P. Hen a d 5, L. Hen y 8, E. an He wijnen 39, M. Heß 64, A. Hicheu 2,
D. Hill 56, M. Hoballah 5, C. Hombach 55, W. Hulsbe gen 42, T. Humai 54, N. Hussain 56, D. Hu chc o 53,
D. Hynds 52, M. Idzik 28, P. Il en 57, R. Jacobsson 39, A. Jaege 12, J. Jalocha 56, E. Jans 42, A. Jawahe y59,
F. Jing 3, M. John 56, D. Johnson 39, C.R. Jones 48, C. Jo am 39, B. Jos 39, N. Ju ik 60, S. Kandybei 44,
W. Kanso 6, M. Ka acson 39, T.M. Ka bach 39,†, S. Ka odia 52, M. Kecke 12, M. Kelsey 60, I.R. Kenyon 46,
M. Kenzie 39, T. Ke el 43, E. Khai ullin 66, B. Khanji 21,39,j, C. Khu ewa hanakul 40, T. Ki n 9, S. Kla e 55,
K. Klimaszewski 29, O. Kochebina 7, M. Kolpin 12, I. Koma o 40, R.F. Koopman 43, P. Koppenbu g 42,39,
M. Kozeiha 5, L. K a chuk 34, K. K eplin 12, M. K eps 49, G. K ocke 12, P. K oko ny 35,69, F. K use 10,
W. K zemien 29, W. Kucewicz 27,n, M. Kucha czyk 27, V. Kud ya se 35,69, A.K. Kuonen40, K. Ku ek 29,
T. K a a skheliya 32, D. Laca e e 39, G. La e y 55,39, A. Lai16, D. Lambe 51, G. Lan anchi 19,
C. Langenb uch 49, B. Langhans 39, T. La ham 49, C. Lazze oni 46, R. Le Gac 6, J. an Lee dam 42, J.-P. Lees 4,
R. Le è e 5, A. Lefla 33,39, J. Le ançois 7, E. Lemos Cid 38, O. Le oy 6, T. Lesiak 27, B. Le e ing on 12, Y. Li 7,
T. Likhomanenko 66,65, M. Liles 53, R. Lindne 39, C. Linn 39, F. Lione o 41, B. Liu 16, X. Liu 3, D. Loh 49,
I. Longs a 52, J.H. Lopes 2, D. Lucchesi 23,q, M. Lucio Ma inez 38, H. Luo 51, A. Lupa o23, E. Luppi 17, ,
O. Lup on 56, A. Lusiani24, F. Mache e 7, F. Maciuc 30, O. Mae 31, K. Magui e 55, S. Malde 56,
A. Malinin 65, G. Manca 7, G. Mancinelli 6, P. Manning 60, A. Mapelli 39, J. Ma a as 5, J.F. Ma chand 4,
U. Ma coni 15, C. Ma in Beni o 37, P. Ma ino 24,39,s, J. Ma ks 12, G. Ma ello i 26, M. Ma in 6,
M. Ma inelli 40, D. Ma inez San os 38, F. Ma inez Vidal 67, D. Ma ins Tos es 2, A. Massa e i 1,
R. Ma e 39, A. Ma had49, Z. Ma he 39, C. Ma euzzi 21, A. Mau i 41, B. Mau in 40, A. Mazu o 46,
M. McCann 54, J. McCa hy 46, A. McNab 55, R. McNul y 13, B. Meadows 58, F. Meie 10, M. Meissne 12,
D. Melnychuk 29, M. Me k 42, E. Michielin 23, D.A. Milanes 63, M.-N. Mina d 4, D.S. Mi zel 12,
J. Molina Rod iguez 61, I.A. Mon oy 63, S. Mon eil 5, M. Mo andin 23, P. Mo awski 28, A. Mo dà 6,
M.J. Mo ello 24,s, J. Mo on 28, A.B. Mo is 51, R. Moun ain 60, F. Muheim 51, D. Mülle 55, J. Mülle 10,
K. Mülle 41, V. Mülle 10, M. Mussini 15, B. Mus e 40, P. Naik 47, T. Nakada 40, R. Nandakuma 50,
A. Nandi56, I. Nas e a 2, M. Needham 51, N. Ne i 22, S. Neube 12, N. Neu eld 39, M. Neune 12,
A.D. Nguyen 40, T.D. Nguyen 40, C. Nguyen-Mau 40,p, V. Niess 5, R. Nie 10, N. Niki in 33, T. Nikodem 12,
A. No oselo 36, D.P. O’Hanlon 49, A. Oblakowska-Mucha28, V. Ob az so 36, S. Ogil y 52,
O. Okh imenko 45, R. Oldeman 16,e, C.J.G. Onde wa e 68, B. Oso io Rod igues 1, J.M. O alo a Goicochea 2,
A. O o 39, P. Owen 54, A. Oyangu en 67, A. Palano14,c, F. Palombo 22, , M. Palu an 19, J. Panman 39,
A. Papanes is 50, M. Pappagallo 52, L.L. Pappala do 17, , C. Pappenheime 58, W. Pa ke 59, C. Pa kes 55,
G. Passale a 18, G.D. Pa el 53, M. Pa el 54, C. Pa ignani 20,i, A. Pea ce55,50, A. Pelleg ino 42, G. Penso 26,l,
M. Pepe Al a elli 39, S. Pe azzini 15,d, P. Pe e 5, L. Pesca o e 46, K. Pe idis 47, A. Pe olini 20,i,
M. Pe uzzo 22, E. Pica os e Olloqui 37, B. Pie zyk 4, T. Pilaˇ
49, D. Pinci 26, A. Pis one20, A. Piucci 12,
S. Play e 51, M. Plo Casasus 38, T. Poikela 39, F. Polci 8, A. Poluek o 49,35, I. Polyako 32, E. Polyca po 2,
A. Popo 36, D. Popo 11,39, B. Popo ici 30, C. Po e a 2, E. P ice 47, J.D. P ice 53, J. P iscianda o 38,
A. P i cha d 53, C. P ou e 47, V. Puga ch 45, A. Puig Na a o 40, G. Punzi 24, , W. Qian 4, R. Quagliani 7,47,
B. Rachwal 27, J.H. Rademacke 47, M. Rama 24, M.S. Rangel 2, I. Raniuk 44, N. Rauschmay 39, G. Ra en 43,
F. Redi 54, S. Reiche 55, M.M. Reid 49, A.C. dos Reis1, S. Riccia di 50, S. Richa ds 47, M. Rihl 39,
K. Rinne 53,39, V. Ri es Molina 37, P. Robbe 7,39, A.B. Rod igues 1, E. Rod igues 55, J.A. Rod iguez Lopez 63,
566 LHCb Collabo a ion / Physics Le e s B 757 (2016) 558–567
P. Rod iguez Pe ez 55, S. Roise 39, V. Romano sky 36, A. Rome o Vidal 38, J.W. Ronayne 13, M. Ro ondo 23,
J. Rou ine 40, T. Ru 39, P. Ruiz Valls 67, J.J. Sabo ido Sil a 38, N. Sagido a 31, P. Sail 52, B. Sai a 16,e,
V. Salus ino Guima aes 2, C. Sanchez Mayo domo 67, B. Sanma in Sedes 38, R. San acesa ia 26,
C. San ama ina Rios 38, M. San ima ia 19, E. San o e i 25,k, A. Sa i19,l, C. Sa iano 26,m, A. Sa a 25,
D.M. Saunde s 47, D. Sa ina 32,33, S. Schael 9, M. Schille 39, H. Schindle 39, M. Schlupp 10,
M. Schmelling 11, T. Schmelze 10, B. Schmid 39, O. Schneide 40, A. Schoppe 39, M. Schubige 40,
M.-H. Schune 7, R. Schwemme 39, B. Sciascia 19, A. Sciubba 26,l, A. Semenniko 32, N. Se a 41, J. Se ano 6,
L. Ses ini 23, P. Sey e 21, M. Shapkin 36, I. Shapo al 17,44, , Y. Shcheglo 31, T. Shea s 53,
L. Shekh man 35,69, V. She chenko 65, A. Shi es 10, B.G. Siddi 17, R. Sil a Cou inho 41, L. Sil a de Oli ei a 2,
G. Simi 23, M. Si endi 48, N. Skidmo e 47, T. Skwa nicki 60, E. Smi h 56,50, E. Smi h 54, I.T. Smi h 51,
J. Smi h 48, M. Smi h 55, H. Snoek 42, M.D. Sokolo 58,39, F.J.P. Sole 52, F. Soom o 40, D. Souza 47,
B. Souza De Paula 2, B. Spaan 10, P. Sp adlin 52, S. S idha an 39, F. S agni 39, M. S ahl 12, S. S ahl 39,
S. S e ko a 54, O. S einkamp 41, O. S enyakin 36, S. S e enson 56, S. S oica 30, S. S one 60, B. S o aci 41,
S. S acka 24,s, M. S a iciuc 30, U. S aumann 41, L. Sun 58, W. Su cli e 54, K. Swien ek 28, S. Swien ek 10,
V. Sy opoulos 43, M. Szczekowski 29, T. Szumlak 28, S. T’Jampens 4, A. Taydugano 6, T. Tekampe 10,
M. Teklishyn 7, G. Tella ini 17, , F. Teube 39, C. Thomas 56, E. Thomas 39, J. an Tilbu g 42, V. Tisse and 4,
M. Tobin 40, J. Todd 58, S. Tolk 43, L. Tomasse i 17, , D. Tonelli 39, S. Topp-Joe gensen 56, N. To 56,
E. Tou nefie 4, S. Tou neu 40, K. T abelsi 40, M.T. T an 40, M. T esch 41, A. T iso ic 39, A. Tsa ego od se 6,
P. Tsopelas 42, N. Tuning 42,39, A. Ukleja29, A. Us yuzhanin 66,65, U. Uwe 12, C. Vacca 16,39,e, V. Vagnoni 15,
G. Valen i 15, A. Vallie 7, R. Vazquez Gomez 19, P. Vazquez Reguei o 38, C. Vázquez Sie a 38, S. Vecchi 17,
M. an Veghel 43, J.J. Vel huis 47, M. Vel i 18,g, G. Veneziano 40, M. Ves e inen 12, B. Viaud 7,∗, D. Viei a 2,
M. Viei es Diaz 38, X. Vilasis-Ca dona 37,o, V. Volko 33, A. Vollha d 41, D. Volyanskyy 11, D. Voong 47,
A. Vo obye 31, V. Vo obye 35, C. Voß 64, J.A. de V ies 42, R. Waldi 64, C. Wallace 49, R. Wallace 13,
J. Walsh 24, S. Wande no h 12, J. Wang 60, D.R. Wa d 48, N.K. Wa son 46, D. Websdale 54, A. Weiden41,
M. Whi ehead 49, G. Wilkinson 56,39, M. Wilkinson 60, M. Williams 39, M.P. Williams 46, M. Williams 57,
T. Williams 46, F.F. Wilson 50, J. Wimbe ley 59, J. Wishahi 10, W. Wislicki 29, M. Wi ek 27, G. Wo mse 7,
S.A. Wo on 48, S. W igh 48, K. Wyllie 39, Y. Xie 62, Z. Xu 40, Z. Yang 3, J. Yu 62, X. Yuan 35,
O. Yushchenko 36, M. Zangoli 15, M. Za e yae 11,b, L. Zhang 3, Y. Zhang 3, A. Zhelezo 12, A. Zhokho 32,
L. Zhong 3, V. Zhuko 9, S. Zucchelli 15
1Cen o B asilei o de Pesquisas Físicas (CBPF), Rio de Janei o, B azil
2Uni e sidade Fede al do Rio de Janei o (UFRJ), Rio de Janei o, B azil
3Cen e o High Ene gy Physics, Tsinghua Uni e si y, Beijing, China
4LAPP, Uni e si é Sa oie Mon -Blanc, CNRS/IN2P3, Annecy-Le-Vieux, F ance
5Cle mon Uni e si é, Uni e si é Blaise Pascal, CNRS/IN2P3, LPC, Cle mon -Fe and, F ance
6CPPM, Aix-Ma seille Uni e si é, CNRS/IN2P3, Ma seille, F ance
7LAL, Uni e si é Pa is-Sud, CNRS/IN2P3, O say, F ance
8LPNHE, Uni e si é Pie e e Ma ie Cu ie, Uni e si é Pa is Dide o , CNRS/IN2P3, Pa is, F ance
9I. Physikalisches Ins i u , RWTH Aachen Uni e si y, Aachen, Ge many
10 Fakul ä Physik, Technische Uni e si ä Do mund, Do mund, Ge many
11 Max-Planck-Ins i u ü Ke nphysik (MPIK), Heidelbe g, Ge many
12 Physikalisches Ins i u , Rup ech -Ka ls-Uni e si ä Heidelbe g, Heidelbe g, Ge many
13 School o Physics, Uni e si y College Dublin, Dublin, I eland
14 Sezione INFN di Ba i, Ba i, I aly
15 Sezione INFN di Bologna, Bologna, I aly
16 Sezione INFN di Caglia i, Caglia i, I aly
17 Sezione INFN di Fe a a, Fe a a, I aly
18 Sezione INFN di Fi enze, Fi enze, I aly
19 Labo a o i Nazionali dell’INFN di F asca i, F asca i, I aly
20 Sezione INFN di Geno a, Geno a, I aly
21 Sezione INFN di Milano Bicocca, Milano, I aly
22 Sezione INFN di Milano, Milano, I aly
23 Sezione INFN di Pado a, Pado a, I aly
24 Sezione INFN di Pisa, Pisa, I aly
25 Sezione INFN di Roma To Ve ga a, Roma, I aly
26 Sezione INFN di Roma La Sapienza, Roma, I aly
27 Hen yk Niewodniczanski Ins i u e o Nuclea Physics Polish Academy o Sciences, K aków, Poland
28 AGH – Uni e si y o Science and Technology, Facul y o Physics and Applied Compu e Science, K aków, Poland
29 Na ional Cen e o Nuclea Resea ch (NCBJ), Wa saw, Poland
30 Ho ia Hulubei Na ional Ins i u e o Physics and Nuclea Enginee ing, Bucha es -Magu ele, Romania
31 Pe e sbu g Nuclea Physics Ins i u e (PNPI), Ga china, Russia
32 Ins i u e o Theo e ical and Expe imen al Physics (ITEP), Moscow, Russia
33 Ins i u e o Nuclea Physics, Moscow S a e Uni e si y (SINP MSU), Moscow, Russia
34 Ins i u e o Nuclea Resea ch o he Russian Academy o Sciences (INR RAN), Moscow, Russia