hep-ex0610039/main.tex
1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %%  ws-procs10x7.tex   :   23 January 2003                               %%
3: %%  Text file to use with ws-procs10x7.cls written in Latex2e.           %%
4: %%  The content, structure, format and layout of this style file is the  %%
5: %%  property of World Scientific Publishing Co. Pte. Ltd.                %%
6: %%  Copyright 1995, 2003 by World Scientific Publishing Co.              %%
7: %%  All rights are reserved.                                             %%
8: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
9: 
10: %\documentclass[draft]{ws-procs10x7}
11: \documentclass{ws-procs10x7}
12: \usepackage{balance}
13: \usepackage{epsfig}
14: 
15: %new column type
16: \newcolumntype{d}[1]{D{.}{.}{#1}}
17: 
18: % A useful Journal macro
19: \def\Journal#1#2#3#4{{\it #1} {\bf #2}, #3 (#4)}
20: 
21: 
22: % -- special symbols
23: \def\rmm   {\ensuremath{\Delta R(\mu\mu)}}
24: \def\mll   {\ensuremath{m_{\mu\mu}}}
25: 
26: 
27: %% -- All numbers
28: \def\vdef #1{\expandafter\def\csname #1\endcsname}
29: \def\vuse #1{\csname #1\endcsname}
30: 
31: \vdef{cut:default:ptlo}    {\ensuremath{  4.0 } } 
32: \vdef{cut:default:pthi}    {\ensuremath{100.0 } } 
33: \vdef{cut:default:etalo}    {\ensuremath{ -2.4 } } 
34: \vdef{cut:default:etahi}    {\ensuremath{  2.4 } } 
35: \vdef{cut:default:rmmlo}    {\ensuremath{  0.3 } } 
36: \vdef{cut:default:rmmhi}    {\ensuremath{  1.2 } } 
37: \vdef{cut:default:tip}      {\ensuremath{0.100 } } 
38: \vdef{cut:default:ptbs}    {\ensuremath{  5.0 } } 
39: \vdef{cut:default:chi2}    {\ensuremath{  1.0 } } 
40: \vdef{cut:default:l3d}    {\ensuremath{0.015 } } 
41: \vdef{cut:default:lxy/sxy}    {\ensuremath{18.0 } } 
42: \vdef{cut:default:lxy}    {\ensuremath{-0.100 } } 
43: \vdef{cut:default:cosalpha}    {\ensuremath{0.995 } } 
44: \vdef{cut:default:isoveto}    {\ensuremath{1000.0 } } 
45: \vdef{cut:default:isolation}    {\ensuremath{0.85 } } 
46: \vdef{cut:default:isocone}    {\ensuremath{  1.0 } } 
47: 
48: 
49: \vdef{mBgSigma:s0}{\ensuremath {36.0 } }
50: \vdef{mBg1s:s0}  {\ensuremath {32.1 } }
51: \vdef{mBg1n:s0} {\ensuremath {0.08 } }
52: \vdef{mBg2s:s0}  {\ensuremath {60.2 } }
53: \vdef{mBg2n:s0} {\ensuremath {0.03 } }
54: 
55: 
56: \vdef{eAllCutsFact:s0}  {\ensuremath{{0.019 } } }
57: \vdef{eAllCutsFactE:s0} {\ensuremath{{0.002 } } }
58: \vdef{nAllCutsFact:s0}  {\ensuremath{{  6.1} } }
59: \vdef{nAllCutsFactE:s0}  {\ensuremath{{  0.3} } }
60: 
61: \vdef{nMassAllCutsFact:m0}  {\ensuremath{{ 13.8} } }
62: \vdef{nMassAllCutsFactE:m0} {\ensuremath{{ 22.0} } }
63: \vdef{eAllCutsFact:m0}  {\ensuremath{{2.74\times 10^{-7} } } }
64: \vdef{eAllCutsFactE:m0}  {\ensuremath{{2.96\times 10^{-8} } } }
65: 
66: \vdef{Lumi:d0}    {\ensuremath{ {10.0 } } }
67: 
68: \vdef{bgError}    {\ensuremath{{160 } } }
69: \vdef{sgError}    {\ensuremath{{25 } } }
70: 
71: 
72: \vdef{nMassAllCutsFactE:s0} {\ensuremath{{ 0.6} } }
73: \vdef{nMassAllCutsFactE:m0} {\ensuremath{{ 22.0} } }
74: 
75: 
76: \vdef{ExpectedUpperLimit}  {\ensuremath{{1.4\times 10^{-8}} } }
77: 
78: 
79: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
80: %                      BEGINNING OF TEXT                                %
81: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
82: \input symbols.tex
83: \makeindex
84: \begin{document}
85: 
86: 
87: \title{STUDY OF \bsmm\ in CMS}
88: 
89: \author{Urs~Langenegger$^*$ }
90: 
91: \address{Institute for Particle Physics, ETH Zurich, 8093 Zurich, Switzerland\\$^*$E-mail: ursl@phys.ethz.ch}
92: 
93: 
94: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
95: % You may repeat \author \address as often as necessary                 %
96: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
97: 
98: \twocolumn[\maketitle\abstract{We  present  a  Monte Carlo  simulation
99:   study  of measuring  the rare  leptonic  decay \bsmm\  with the  CMS
100:   experiment  at the  LHC.   The study  is  based on  a full  detector
101:   simulation  for  signal  and  background  events.   We  discuss  the
102:   high-level  trigger algorithm  and the  offline event  selection.  }
103: \keywords{Rare $B$ decays; Supersymmetry; LHC} ]
104: 
105: 
106: 
107: % ----------------------------------------------------------------------
108: \section{Introduction}
109: 
110: In the standard  model (SM), the decay \bsmm\  has a highly suppressed
111: rate\cite{Buras:2003td} of $\cbf  = (3.42\pm0.54)\times 10^{-9}$ since
112: it  involves a $b  \to s$  transition and  requires an  internal quark
113: annihilation  which  further  suppresses  the decay  relative  to  the
114: electroweak `penguin' $b \to s \gamma$ decay.  In addition, the decays
115: are helicity  suppressed by factors of $m_{\ell}^{2}$.   To date these
116: decays  have  not  been  observed;  upper limits  on  these  branching
117: fractions    are    a   topic    of    frequent    updates   at    the
118: $B$-factories\cite{Chang:2003yy,Aubert:2004gm} (for the decay $\bdmm$)
119: and  the Tevatron.\cite{Abazov:2004dj,CDF:prelim}  Currently  the best
120: upper  limit  is  from  the  CDF  collaboration\cite{CDF:prelim}  with
121: $\cbf(\bsmm) < 1.0\times 10^{-7}$ at 95\% confidence limit.
122: 
123: Since  these processes  are  highly  suppressed in  the  SM, they  are
124: potentially   sensitive  probes   of  physics   beyond  the   SM  (see
125: Fig.~\ref{f:b2llnp}).  In the  minimal supersymmetric extension of the
126: SM (MSSM) the branching fraction for these decays can be substantially
127: enhanced, especially at large $\tan \beta$.\cite{Babu:1999hn} For MSSM
128: with  modified  minimal flavor  violation  at  large $\tan\beta$,  the
129: branching  fraction  can  be  increased   by  up  to  four  orders  of
130: magnitude.\cite{Bobeth:2002ch} \bsdmm\ decays  can also be enhanced in
131: specific  models   containing  leptoquarks~\cite{Davidson:1993qk}  and
132: supersymmetric (SUSY) models without R-parity.\cite{Roy:1991qr}
133: 
134: \begin{figure}[!htb]
135:  \centerline{\psfig{file=fig1.eps,width=2.6in}}
136:  \caption{Feynman graphs for the decay \bsdmm\ illustrating possible
137:    new physics contributions.}
138:  \label{f:b2llnp}
139: \end{figure}
140: 
141: There                  has                  been                  some
142: interest\cite{Kane:2003wg,Baek:2004tm,Dedes:2004yc} in using the decay
143: mode \bsmm\  to `measure' the  key parameter \tb\  of the MSSM  and to
144: constrain other  extensions of  the SM.  Lower  bounds on \tb\  can be
145: obtained    from   $\cbf(\bsmm)$   with    general   model-independent
146: assumptions.  Since \tb\ is also constrained from above due to general
147: principles, a lower bound is tantamount to a measurement of \tb.
148: 
149: 
150: 
151: % ----------------------------------------------------------------------
152: \section{The CMS Experiment}
153: 
154: The Compact Muon Solenoid (CMS)  detector is well suited for the study
155: of leptonic $B$ decays. The  main components for this analysis are the
156: tracker and muon systems.
157: 
158: The CMS tracker  is an all-silicon detector, comprises  an inner pixel
159: vertex detector plus an outer strip track detector, and is immersed in
160: a   magnetic  field   of   4\Tesla.   The   pixel  detector   provides
161: high-precision measurements  of space points close  to the interaction
162: region  for  vertexing  and   effective  pattern  recognition  in  the
163: high-track multiplicity environment at  the LHC. The pixel detector is
164: composed of 1440 modules arranged  in three barrel layers (at a radial
165: distance of  $r = 4.4,\, 7.3,\,  10.2\cm$ from the  beampipe) and four
166: endcap  disks (at  $z =  \pm34.5,\, \pm46.5\cm$  from  the interaction
167: point).    The  barrel   detector   comprises  672   modules  and   96
168: half-modules,  the forward  detector is  built of  96 blades  with 672
169: modules.   With a  pixel  size of  $d_\phi\times  d_z =  100\mum\times
170: 150\mum$,  a hit  resolution of  10--20\mum\ is  achieved.   The strip
171: tracker allows the precise  determination of charged particle momenta. 
172: It is  arranged in  the central  part as inner  (TIB) and  outer (TOB)
173: barrel, and  in the forward regions  as inner discs  (TID) and endcaps
174: (TEC).  The barrel part consists of 10  layers (4 in the TIB, 6 in the
175: TOB) and 12 layers  in the forward part (3 in the TID,  9 in the TEC). 
176: The  pitch  in the  strip  tracker  varies  between 80--180\mum.   The
177: material  inside  the active  volume  of  the  tracker increases  from
178: $\approx 0.4X_0$  at pseudo-rapidity  $\eta = 0$  to around  $1X_0$ at
179: $|\eta| \approx 1.6$, before decreasing to $\approx 0.6X_0$ at $|\eta|
180: = 2.5$.
181: %The  transverse  momentum
182: %resolution of the tracker can be approximately parametrized as $\Delta
183: %\pt/\pt = 0.02\%\pt/\gev \oplus 0.5\%$.
184: 
185: The  CMS muon  system, incorporated  into the  magnet return  yoke, is
186: divided  into a  barrel ($|\eta|  < 1.2$)  and forward  parts  ($1.2 <
187: |\eta|  < 2.4$).   In the  barrel  region, where  the neutron  induced
188: background and  the muon rate is  small, drift tube  (DT) chambers are
189: used. In  the two endcaps cathode  strip chambers (CSC)  are deployed. 
190: In  addition, resistive  plate chambers  (RPC)  are used  both in  the
191: barrel and the  endcap region.  The RPC spatial  resolution is coarser
192: than for  the DT and CSC,  but their excellent  time resolution allows
193: the unambiguous identification of the correct bunch crossing.
194: 
195: 
196: 
197: 
198: % ----------------------------------------------------------------------
199: \section{Event Samples}
200: 
201: Monte Carlo (MC) event samples were generated with {\sc Pythia~6.227}
202: and passed through a full detector  simulation based on {\sc Geant~4}. 
203: On  average  five pile-up  events  were  included,  appropriate for  a
204: luminosity of $\clu = 2\times10^{33} \cm^{-2}\sec^{-1}$.
205: 
206: Both signal  and background  MC event samples  have been  generated as
207: minimum bias  QCD events. In the  signal sample, \Bs\  mesons decay as
208: \bsmm.  The muons are required  to have transverse momentum $\pt^\mu >
209: 3\gev$ and $|\eta^\mu| < 2.4$; the \Bs\ must have $\pt^{\Bs} > 5\gev$.
210: 
211: The  background sample  contains  two  muons with  $\pt  > 3\gev$  and
212: $|\eta|  <  2.4$.   Their  separation in  azimuth  and  pseudorapidity
213: $\rmm\equiv \sqrt{\Delta\phi^2 + \Delta\eta^2}$ is required to be $0.3
214: < \rmm <  1.8$.  Currently the background simulation  does not include
215: muons  due  to hadronic  in-flight  decays  or  punch-through.  It  is
216: estimated that  this hadronic  component will increase  the background
217: level    by    about    10\%.     Background    events    from    rare
218: $B_d,\,B_u,\,B_s,\,B_c,\,\Lambda_b$ decays are not included.
219: 
220: In total  20000 signal events  and about 15000 background  events have
221: been  analyzed.   The small  size  of  the  background sample  is  the
222: limiting factor of the present study.
223: 
224: 
225: 
226: % ----------------------------------------------------------------------
227: \section{Trigger Strategy}
228: 
229: The  CMS detector  has a  twofold trigger  strategy.  The  first level
230: trigger,  with a latency  of $3.2\,\mu\sec$,  is based  on information
231: from  the  calorimeters  and  the  muon  system.   The  threshold  for
232: inclusive  isolated single  muons is  at $\pt^\mu>14\gev$,  for dimuon
233: events both  muons must have $\pt>3\gev$.  The  expected trigger rates
234: amount to  2.7\kHz\ and 0.9\kHz, respectively.  The  L1 trigger output
235: rate is at most 100\kHz.
236: 
237: The high-level trigger (HLT) is a software trigger, running on a large
238: processor  farm. The  HLT reduces  the overall  trigger rate  by three
239: orders  of magnitude  to about  100\Hz.  To  fit into  the  tight time
240: constraints imposed  by the  high input rate,  tracking at the  HLT is
241: sped up by  two concepts: (1) `Regional seed  generation' limits track
242: seeding to  specific regions  of interest, \eg,  a cone around  the L1
243: muon  candidate  direction.   (2)  `Partial  tracking'  pursues  track
244: reconstruction only until some criteria  are met, \eg, a resolution of
245: 2\% on  the transverse momentum. Already with  six reconstructed hits,
246: both  the efficiency  and the  resolution are  comparable to  the full
247: tracking performance.
248: 
249: The HLT  strategy for  \bsmm\ proceeds along  the following  path. (1)
250: Verification of the two  L1 muon candidates.  (2) Tracks reconstructed
251: only with  the pixel detector are  used to compute a  list of possible
252: primary  vertices,  the  three  most significant  are  retained.   (3)
253: Regional  track reconstruction  with up  to six  hits is  performed in
254: cones around  the L1 muon  candidates.  (4) Reconstructed  tracks with
255: $\pt >  4\gev$ are paired and  retained if their  invariant mass falls
256: into predefined  regions for  the signal and  sidebands.  (5)  The two
257: tracks must have  opposite charge and are fit to  a common vertex; the
258: event  is  retained   only  when  the  fit  $\chi^2   <  20$  and  the
259: three-dimensional  flight  length   $l_{3d}  >  150\mum$.   With  this
260: selection,  the event  rate was  estimated\cite{Sphicas:2002gg}  to be
261: $<1.7\Hz$.
262: 
263: 
264: % ----------------------------------------------------------------------
265: \section{Offline Analysis}
266: 
267: The offline analysis  selection focuses on a secondary  vertex that is
268: well  measured and separated  from the  primary vertex  and consistent
269: with the decay of an isolated \Bs\ meson.
270: 
271: The primary vertex is determined from all tracks with $\pt > 0.9\gev$,
272: because of the rather low-multiplicity track environment.
273: 
274: The   two    muons   must    have   opposite   charge,    $\pt^\mu   >
275: \vuse{cut:default:ptlo} $,  and $|\eta^\mu| < \vuse{cut:default:etahi}
276: $.   The azimuthal  and  pseudorapidity separation  of  the two  muons
277: $\vuse{cut:default:rmmlo} < \rmm < \vuse{cut:default:rmmhi} $ provides
278: a  powerful  reduction  of  gluon-gluon fusion  background  with  both
279: $b$-hadrons decaying semileptonically:  The muons of those $b$-hadrons
280: tend to be back-to-back, while  the signal shows a peaked distribution
281: with a maximum at $\rmm \sim 1$.
282: 
283: \Bs\ candidates are formed by  vertexing the two muon candidates.  The
284: vertex quality is required to  be $\chi^2 < \vuse{cut:default:chi2} $. 
285: The transverse momentum vector of  the \Bs\ candidate must be close to
286: the displacement of the secondary  vertex from the primary vertex: the
287: cosine  of the  opening angle  $\alpha$ between  the two  vectors must
288: fulfill   $\cos(\alpha)    >   \vuse{cut:default:cosalpha}   $.    The
289: significance  of the  \Bs\  candidate flight  length  $l_{xy}$ in  the
290: transverse  plane is defined  as $l_{xy}/\sigma_{xy}$  (illustrated in
291: Fig.~\ref{f:selection}),  where  $\sigma_{xy}$  is  the error  on  the
292: flight  length  with mean  $\langle\sigma_{xy}\rangle  = 120\mum$;  we
293: require $l_{xy}/\sigma_{xy} > \vuse{cut:default:lxy/sxy} $.
294: 
295: \begin{figure}[!htb]
296: %%% \centerline{\psfig{file=fig3.eps,width=2.6in}}
297:  \centerline{\psfig{file=fig2.eps,width=2.6in}}
298:  \caption{Decay length significance in
299:    the  transverse plane  for  signal and  background  MC events.  Both
300:    histograms are normalized to unity.}
301: %%  \caption{(a) Dimuon isolation and (b) decay length significance measured in
302: %%    the transverse plane for signal and background MC events.}
303:  \label{f:selection}
304: \end{figure}
305: 
306: In  high-\pt\  gluon-splitting events  the  \bbbar\  quark pair  moves
307: closely together due to their boost, and the two decay vertices of the
308: resulting $b$-hadrons cannot be well separated in all cases.  However,
309: because of color reconnection, the hadronic activity around the dimuon
310: direction is  enhanced compared  to the signal  decay (where  only one
311: colorless  \Bs\  meson  decays).    This  is  exploited  in  isolation
312: requirements:  The  isolation  $I$   is  determined  from  the  dimuon
313: transverse momentum  and charged tracks  with $\pt>0.9\gev$ in  a cone
314: with half-radius  $r = \vuse{cut:default:isocone} $  around the dimuon
315: direction as $I \equiv \pt^{\mu\mu}/(\pt^{\mu\mu} + \sum_{trk}|\pt|)$;
316: we require $I > \vuse{cut:default:isolation} $.
317: 
318: 
319: \begin{figure}[!htb]
320:  \centerline{\psfig{file=fig4.eps,width=2.6in}}
321:  \caption{Dimuon mass distribution in signal MC events. The curve
322:    is a  fit of two  Gaussians, the displayed parameters  indicate the
323:    average mean and sigma. The histogram is normalized to unity.}
324:  \label{f:breco}
325: \end{figure}
326: 
327: 
328: Figure~\ref{f:breco} illustrates  the mass resolution  obtained on the
329: signal MC  event sample. The  distribution is fit with  two Gaussians,
330: the  quoted width  $\sigma  = \vuse{mBgSigma:s0}  \mev$ is  determined
331: according to
332: \begin{eqnarray*}
333:  \sigma^2 = \frac{N_n^2\sigma_n^2 + N_w^2\sigma_w^2}{N_n^2 + N_w^2},
334: \end{eqnarray*}
335: where $\sigma_n  = \vuse{mBg1s:s0} \mev $ ($\sigma_w =  \vuse{mBg2s:s0} \mev$)
336: and $N_n =  \vuse{mBg1n:s0} $ ($N_w = \vuse{mBg2n:s0} $)  are the width
337: and normalization of the narrow (wide) Gaussian, respectively.
338: 
339: Given  the  limited statistics  of  the  background  sample, no  event
340: remains after the application  of all selection requirements. However,
341: the absence of correlation  to the other selection requirements allows
342: a factorization  of the isolation  and $\chi^2$ requirements  from the
343: other  cuts in  the determination  of the  total  background rejection
344: factor.    The  dominant   sources  of   uncertainty  on   the  signal
345: ($\pm\vuse{sgError}  \%$) and background  ($^{+\vuse{bgError} }_{-100}
346: \%$) yield are the statistical component of the background sample, the
347: impact   of  the   misalignment  on   the  transverse   flight  length
348: significance, and the assumption of factorizing cuts.
349: 
350: The total  selection efficiency for signal events  is $\varepsilon_S =
351: \vuse{eAllCutsFact:s0}   \pm   \vuse{eAllCutsFactE:s0}   $   and   the
352: background reduction factor is $\varepsilon_B = \vuse{eAllCutsFact:m0}
353: $.   With  this  selection,   the  first  $\vuse{Lumi:d0}  \invfb$  of
354: integrated  luminosity   will  yield  $n_S   =  \vuse{nAllCutsFact:s0}
355: \pm\vuse{nMassAllCutsFactE:s0}   $   signal    events   and   $n_B   =
356: \vuse{nMassAllCutsFact:m0}               ^{+\vuse{nMassAllCutsFactE:m0}
357: }_{-\vuse{nMassAllCutsFact:m0} } $ background  events in a mass window
358: of $m_{\Bs}  \pm 0.1\gev$.  With  this background estimate,  the upper
359: limit    on    the   branching    fraction    is   $\cbf(\bsmm)    \le
360: \vuse{ExpectedUpperLimit} $ at the 90\% C.L.
361: 
362: 
363: 
364: 
365: % ----------------------------------------------------------------------
366: \section{Conclusions}
367: 
368: This study is limited by the size of the background MC sample.  In the
369: future,  it will  include  larger background  samples  and a  detailed
370: simulation of  rare $b$-hadron decays. The search  for \bsmm\ promises
371: an interesting start-up analysis  with the possibility of setting tight
372: constraints on  the MSSM.  With sufficient  integrated luminosity, the
373: precision  measurement  of  the  \bsmm\ branching  fraction  will  set
374: constraints on models of new physics.
375: 
376: 
377: \input references.tex
378: 
379: \end{document}
380: