1: \section{Analysis}\label{sec:analysis}
2: Although the expressions for black hole production given
3: in Section~\ref{sec:intro} are not valid at $\mbh \sim \TPs$,
4: our BH generator applies them in all the regions where
5: $\mbh > \TPs$ in this study.
6:
7: \subsection{Selection Criteria and Reconstruction}\label{sec:criteria}
8: The following cuts are applied in order to select and reconstruct black holes~(BHs):
9: \begin{itemize}
10: \item For a precise reconstruction of the BH, it is necessary to remove particles
11: produced in the stage of initial state radiations~(ISRs).
12: Because the particles from ISRs tend to have small $\pt$ and large $|\eta|$,
13: as shown in Figures.~\ref{fig:pt} and \ref{fig:eta},
14: the following requirements are applied to particles in each event
15: ~(ISR-cut)~:\\ \\
16: \hspace*{0.75cm} $\bullet \hspace{0.25cm} \pt ~>~30~\textrm{GeV} \hspace{0.75cm} \textrm{for} \ \mu, e$ \\
17: \hspace*{0.75cm} $\bullet \hspace{0.25cm} \pt ~>~ 50~\textrm{GeV} \hspace{0.75cm}\textrm{for} \ \gamma,$ jet\\
18:
19: \begin{figure}[H]
20: \begin{center}
21: \resizebox{0.48\textwidth}{!}{\includegraphics{figs/bh-atlas-note-pt3.epsi}}
22: \resizebox{0.50\textwidth}{!}{\includegraphics{figs/bh-atlas-note-pt2.epsi}}
23: \end{center}
24: \caption{$\pt$ distributions~:~comparison between particles from BH and
25: initial state radiation~($\TPs=1$~TeV, $n$=3).}
26: \label{fig:pt}
27: \end{figure}
28:
29:
30: \begin{figure}[H]
31: \begin{center}
32: \resizebox{0.48\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-eta.eps}}
33: \end{center}
34: \caption{$\eta$ distributions~:~comparison between particles from BH and initial state radiation~($\TPs=1$~TeV, $n$=3).}
35: \label{fig:eta}
36: \end{figure}
37:
38:
39: \item Of all the particles passing the ISR-cut,
40: more than three
41: are required to have energy larger than 300~GeV~($E>300$~GeV),
42: and moreover, at least one of them has to be
43: either an electron or a photon.
44: This latter requirement is to suppress backgrounds.
45: Figure~\ref{fig:n-particles} shows the distributions of the number
46: of energetic particles in each event, where \bggj ~event is one of
47: the largest contribution in the backgrounds as shown in Table~\ref{table:bg}.
48: It is seen that this multiplicity cut on energetic particles is very effective.
49:
50: \item We require the event shape variable $R_2$ to be less than 0.8~($R_2<0.8$).
51: $R_2$ represents an event topology and it is defined by
52: Fox-Wolfram moments as follows:
53: \[R_2 \equiv H_2 /H_0\]
54: \[H_i \equiv \sum _{j,k} \frac{\mid p_j ^* \mid
55: \mid p_k ^* \mid}{E^2} P_i
56: (cos \phi _{jk})\]
57:
58: where $H_i$ is the $i$-th Fox-Wolfram moments, $j$ and $k$ are
59: ID numbers of tracks, $p_j^*$ is the momentum of the
60: %track $j$ in the cms, $\phi _{jk}$ is the opening angle between
61: track $j$ in the rest frame of BH, $\phi _{jk}$ is the opening angle between
62: the tracks $j$ and $k$ ,
63: % E is the energy of BH in the cms,
64: and $P_i(x)$ is the Legendre polynomial.
65: E is obtained by summing up the energies of particles passing the ISR-cut,
66: calculated in the rest frame of BH.
67: $R_2$ ranges from 0 to 1 ($0\leq R_2 \leq 1$) as can be seen from
68: Figure~\ref{fig:r2} which shows the $R_2$ distribution for signal and $qq$ events with or without selection criteria.
69: Since lower value of $R_2$ indicates more spherical event,
70: we remove $qq$ events here.
71:
72: \begin{figure}[H]
73: \begin{center}
74: \resizebox{0.48\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-nh.epsi}}
75: \end{center}
76: \caption{Shape of the distribution of the number of energetic particles with $E$ $>$ 300~GeV~:~comparison between signal and \bggj ~background~($\TPs=1$~TeV, $n$=3).}
77: \label{fig:n-particles}
78: \end{figure}
79:
80: \begin{figure}[H]
81: \begin{center}
82: \resizebox{0.48\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-r2-beforecut.eps}}
83: \resizebox{0.48\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-r2.eps}}
84: \end{center}
85: \caption{$R_2$ Distributions~:~comparison between signal event and
86: $qq$~event~($\TPs=1$~TeV, $n$=3).
87: Left: without any selection criteria and
88: right: with selection criteria}
89: \label{fig:r2}
90: \end{figure}
91:
92:
93: %The $R_2$ is defined as a ratio of 2$^\mathrm{nd}$ to 0$^\mathrm{th}$ Fox-Wolfram moments:
94: %\begin{align*}
95: %R_2 &\equiv H_2/H_0,
96: %\end{align*}
97: %where the $i^\mathrm{th}$ Fox-Wolfram moment $H_i$ is defined as
98: %\begin{align*}
99: %H_i &\equiv \sum_{j,k}\frac{|\vec{pS}^*_j||\vec{p}^*_k|}{E^2}P_i(\cos \phi_{jk}),
100: %\end{align*}
101: %where the indices $j,k$ run over all tracks in the event, $\vec{p}^*_j$
102: %is the momentum of the track $j$ in the cms, $E$ is the total
103: %energy of the event in the cms, $\phi_{jk}$ is the angle
104: %between the track $j$ and $k$, and $P_i(x)$ is the $i^\mathrm{th}$ Legendre polynomial\footnote{
105: %$P_0(x) = 1, P_1(x) = x, P_2(x) = (3x^2-1)/2$}.
106:
107:
108:
109: \item $\mET<100$~GeV.
110: %This cut is necessary for the precise calculation of a BH mass.
111: For a precise calculation of BH mass, events with high missing energy
112: are rejected here.
113: \end{itemize}
114:
115:
116: The black hole mass is then reconstructed from the 4-momenta of the remaining muons, electrons, gammas and jets as follows:
117: \[
118: p_{\mathrm{BH}} = \sum_{i=\mu,e,\gamma,jet}p_i
119: \]
120: \[
121: \mbh = \sqrt{p_{\mathrm{BH}}^2},
122: \]
123: where $p_i$ is a reconstructed four-momentum of each particle.
124:
125: The quality of the reconstructed mass of the BH is discussed in Appendix~\ref{app:mass}.
126:
127: \subsection{Discovery Potential}
128:
129: We have evaluated the discovery potential for the following cases of ($\TPs$ ,$n$):
130: \begin{center}
131: $
132: \begin{cases}
133: \TPs =1,3,4,5,6,7~\mathrm{TeV} \\
134: n = 2,3,5,7
135: \end{cases}
136: $
137: \end{center}
138: by using the selection criteria described in Section~\ref{sec:criteria}.
139: Because $\Mp$ is an unknown parameter, we cannot set the proper cut value
140: for the lower limit of $\mbh$.
141: Therefore we consider various values as follows and evaluate
142: $S/\sqrt{B}$ in each case:
143: \begin{center}
144: $\mbh>\mbhmin = 1,2,3,4,5$~TeV.
145: \end{center}
146:
147: The $\mbh$ distributions are shown in Figure~\ref{fig:bh-mass-dist}
148: for different values of $\TPs$ with $n=3$.
149: These are obtained after all the selection criteria have been applied.
150: The histograms represent the sum of signals and backgrounds
151: with the cross-hatched part showing only backgrounds.
152: Note that $\mbhmin$ is set to 1~TeV for this figure.
153:
154:
155: \begin{figure}[H]
156: \hspace{1.2cm} (a) $\Mp$=1~TeV \hspace{5.0cm} (b) $\Mp$=3~TeV
157: \vspace{-1.0cm}
158: \begin{center}
159: \resizebox{0.43\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-mbh-131.epsi}}
160: \resizebox{0.43\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-mbh-331.epsi}}
161: \end{center}
162: \hspace{1.2cm} (c) $\Mp$=4~TeV \hspace{5.0cm} (d) $\Mp$=5~TeV
163: \vspace{-1.0cm}
164: \begin{center}
165: \resizebox{0.43\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-mbh-431.epsi}}
166: \resizebox{0.43\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-mbh-531.epsi}}
167: \end{center}
168: \hspace{1.2cm} (e) $\Mp$=6~TeV \hspace{5.0cm} (f) $\Mp$=7~TeV
169: \vspace{-1.0cm}
170: \begin{center}
171: \resizebox{0.43\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-mbh-631.epsi}}
172: \resizebox{0.43\textwidth}{!}{\includegraphics{figs/bh-atlas-note-lboff-mbh-731.epsi}}
173: \end{center}
174: \caption{$\mbh$ distributions in the case of $n$=3 and $\mbhmin$=1~TeV.~(solid line~$:$~ signal plus background, cross hatched~$:$~background only)}
175: \label{fig:bh-mass-dist}
176: \end{figure}
177:
178:
179: Tables~\ref{table:bg}-\ref{table:signal-event-number2} show
180: the number of events passing the selection criteria for background~(BG)
181: processes and signal events with various ($\TPs$ ,$n$).
182: The number is normalized for \ILu =10~\Fb.
183: The values of efficiency are listed only for $\mbhmin = 1$~TeV.
184: As can be seen in Table~\ref{table:bg}, \bggj~process has the
185: highest efficiency, but the background is dominated by the
186: $qq$ process owing to its large cross section.
187: The total number of BG events is found to be
188: $\sim 10^3$~-~$10^4$ in the case of $\mbhmin = $~1~-~2~TeV
189: and $\sim 10$~-~$10^2$ in the case of
190: $\mbhmin = $~4~-~5~TeV.
191:
192: According to Tables~\ref{table:signal-event-number1} and
193: \ref{table:signal-event-number2}, the efficiency for signal events
194: is $\sim$ 1\%
195: in all cases of ($\TPs$ ,$n$).
196: It is relatively lower
197: when $\TPs = 1$~TeV or $7$~TeV
198: because signal events are more likely to be rejected by the $\mbhmin$~cut
199: in the case of smaller $\TPs$, or they have a large $\mET$ for larger
200: $\TPs$.
201:
202: \input{ana-table-number}
203:
204: \input{ana-table-sb}
205:
206: We calculate $S/\sqrt{B}$ from the data of Tables~\ref{table:bg}-
207: \ref{table:signal-event-number2} and estimate
208: \ILuD~---~the integrated luminosity
209: with which discovery is achieved.
210: Here the condition of discovery is set as:
211: \[S/\sqrt{B}~ \geq ~ 5.0
212: \hspace{0.75cm}
213: \textrm{and}
214: \hspace{0.75cm}
215: S~\geq ~10 \]
216: which is a conventional condition, as used in the analysis of Higgs events
217: with ATLAS.
218: The results for $S/\sqrt{B}$ and \ILuD\
219: are shown in Table~\ref{table:soverrootb1}-\ref{table:soverrootb2}
220: and \ref{table:ldiscovery} respectively.
221: We require that $\mbhmin$ be larger than $\TPs$ to calculate
222: integrated luminosities for BH discovery.
223: In Table~\ref{table:ldiscovery}, the shaded values
224: indicate the most favorable cut in $\mbhmin$
225: in each case of ($\TPs$ ,$n$).
226: From these tables, we see that the discovery can be accomplished
227: within \ILuD~$\le~1$~\Fb~ in all cases of
228: $n$ if $\TPs$ is less than $\sim$ 5~TeV.
229:
230: \input{ana-table-ldis}
231:
232: Figure~\ref{fig:contours} gives a contour plot
233: for \ILuD\ in ($\TPs$ ,$n$) plane.
234: We find that the discovery potential hardly depends on $n$
235: but has a strong dependence on $\TPs$.
236: This is due to the fact that the
237: cross section is a strong function of $\TPs$
238: but not of $n$,
239: as shown in Tables~\ref{table:signal-event-number1} and \ref{table:signal-event-number2}.
240: From this relation between $\TPs$ and \ILuD,
241: $\TPs$ can be determined by how early the discovery is accomplished.
242:
243: \begin{figure}[H]
244: \begin{center}
245: \resizebox{0.70\textwidth}{!}{\includegraphics{figs/bh-atlas-note-contour-c.epsi}}
246: %\resizebox{0.70\textwidth}{!}{\includegraphics{figs/bh-atlas-note-contour.epsi}}
247: %\includegraphics[height=10cm,width=10cm,clip]{figs/bh-atlas-note-contour.epsi}
248: \end{center}
249: \caption{Contours of \ILuD\ in ($\TPs$,~$n$) plane. }
250: \label{fig:contours}
251: \end{figure}
252:
253:
254: In a quantitative point of view,
255: it is found that the excess of events is detected
256: in $\sim$ 1~month at low luminosity~(\ILu$=$1~\Fb) if $\TPs < 5$~TeV,
257: and discovery within only one day~(\ILu =100~\Pb)
258: can be expected if $\TPs < 4$~TeV.
259:
260: %In the above study, we assume that the expressions of the black hole
261: %in Section~\ref{sec:intro}
262: %are valid in the all region of $\mbh>\TPs$.
263: %Strictly speaking, they are valid in the $\mbh\gg\TPs$ and
264: %it is difficult to say what happens in the $\mbh<\sim5\TPs$.
265:
266: As was previously mentioned, the BH model we assumed here is valid
267: only when $\mbh \gg \TPs$.
268: As $\mbh$ approaches $\TPs$, the theory of BH production
269: becomes very complex.
270: %Actually it is difficult to say
271: %what happens in the case of $\mbh<\sim5\TPs$.
272: If we consider events with reconstructed $\mbh>5$~TeV in the case of
273: $\TPs=1$~TeV, we find that we need $\sim$5~\Pb\ instead of $\sim$0.1~\Pb\
274: to discover black holes.
275: However the events include black holes
276: whose generated mass was less then 5~TeV.
277: When black holes with mass greater than 5~TeV are generated, we find that we need $\sim$10~\Pb.
278:
279: Considering that the model of BH formation and decay is valid only for
280: $\mbh \gg \TPs$, we show, in Figure~\ref{fig:contours2}, an evaluation of the discovery potential
281: of BHs when the cut $\mbhmin > \TPs +1$~TeV is applied.
282: Although more integrated luminosity is required, it is clear that an excess of events will
283: still be easily observed within a few days of running, for $\TPs$ values up to a few TeV.
284:
285: \begin{figure}[H]
286: \begin{center}
287: \resizebox{0.70\textwidth}{!}{\includegraphics{figs/bh-atlas-note-contour-b.epsi}}
288: \end{center}
289: \caption{Contours of \ILuD\ in ($\TPs$,~$n$) plane in case of $\mbhmin > \TPs +1$~TeV.}
290: \label{fig:contours2}
291: \end{figure}
292: