1: %\documentclass[12pt]{article}
2: \documentclass{cernrep}
3: \usepackage{cite,epsfig,amssymb,amsmath}
4: \bibliographystyle{JHEP}
5:
6:
7: \newlength{\fdagwidth}
8: \newlength{\diagupwidth}
9: \newlength{\stepback}
10: \newcommand{\fdag}[2][\diagup]{\text{$#2$\settowidth{\fdagwidth}{$#2$}\settowidth{\diagupwidth}{$#1$}\setlength{\stepback}{0.5\fdagwidth}\hspace{-\stepback}\hspace{-0.5\diagupwidth}$#1$\hspace{\stepback}\hspace{-0.5\diagupwidth}}}
11:
12:
13: \begin{document}
14: \begin{titlepage}
15: \nopagebreak
16: {\flushright{
17: \begin{minipage}{5cm}
18: \Large KA--TP--05--2006 \\
19: {\tt hep-ph/0605117}\hfill \\
20: \end{minipage} }
21:
22: }
23: \vfill
24: \begin{center}
25: {\LARGE \bf \sc
26: \baselineskip 0.9cm
27: Higgs + 2 Jets as a Probe for CP
28: Properties~\footnote{Contribution to the Proceedings of the Workshop
29: on CP Violation and Non-standard Higgs Physics, CERN}
30: }
31: \vskip 0.5cm
32: {\large
33: V.~Hankele, G.~Kl\"amke, and D.~Zeppenfeld
34: }
35: \vskip .2cm
36: {\it Institut f\"ur Theoretische Physik,
37: Universit\"at Karlsruhe, P.O.Box 6980, 76128 Karlsruhe, Germany}
38:
39:
40: \vskip
41: 1.3cm
42: \end{center}
43:
44: \nopagebreak
45: %\vfill
46: %\vskip 3cm
47: \begin{abstract}
48: Azimuthal angle correlations of the jets in Hjj events at the LHC provide a
49: probe of the CP nature of Higgs couplings to gauge bosons. In weak boson
50: fusion the HWW and HZZ couplings are tested. Gluon fusion processes probe
51: the tensor structure of the effective Hgg vertex and thus the CP
52: nature of the dominant quark couplings.
53: \end{abstract}
54: \vfill
55: %\today \timestamp \hfill
56: \vfill
57:
58: % PACS: 14.80.Bn
59: \end{titlepage}
60: %
61: %\newpage
62: %
63: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
64: %
65:
66:
67:
68: %---------------------------------------------------------------------
69:
70: %% \title{Higgs + 2 Jets as a Probe for CP
71: %% Properties%~\footnote{Contribution to the Proceedings of the Workshop
72: %% % on CP Violation and Non-standard Higgs Physics, CERN}
73: %% }
74:
75: %% \author{V.~Hankele, G.~Kl\"amke, and D.~Zeppenfeld}
76: %% \institute{Institut f\"ur Theoretische Physik,
77: %% Universit\"at Karlsruhe, P.O.Box 6980, 76128 Karlsruhe, Germany}
78:
79: %% \maketitle
80:
81: %% \begin{abstract}
82: %% Azimuthal angle correlations of the jets in Hjj events provide a
83: %% probe of the CP nature of Higgs couplings to gauge bosons. In weak boson
84: %% fusion the HWW and HZZ couplings are tested. Gluon fusion processes probe
85: %% the tensor structure of the effective Hgg vertex and thus the CP
86: %% nature of the dominant quark couplings.
87: %% \end{abstract}
88:
89:
90: %Contact: dieter@particle.uni-karlsruhe.de\\
91:
92: \noindent
93: At the LHC, one would like to experimentally determine the CP nature of
94: any previously discovered (pseudo)scalar resonance. Such measurements
95: require a complex event structure in order to provide the
96: distributions and correlations which can distinguish between CP-even and
97: CP-odd couplings. This can either be done by considering
98: decays, e.g. $H\to ZZ\to l^+l^- l^+l^-$ and the correlations of the
99: decay leptons~\cite{Choi:2002jk,Buszello:2002uu},
100: or one can study correlations arising in the production
101: process. Here the azimuthal angle correlations between the two
102: additional jets in $Hjj$ events have emerged as a promising
103: tool~\cite{Plehn:2001nj}. In the following we consider the prospects for
104: using $\Phi jj$ events at the LHC, where $\Phi$ stands for either a CP
105: even boson, H, or a CP odd state, A. Two production processes are considered.
106: The first is vector boson fusion (VBF), i.e. the electroweak process $qQ\to
107: qQ\Phi$ (and crossing related ones) where $\Phi$ is radiated off a
108: $t$-channel electroweak boson. The second is gluon fusion where $\Phi$ is
109: produced in QCD dijet events, via the insertion of a heavy quark loop
110: which mediates $gg\to\Phi+0,1,2$ gluons.
111:
112: The CP properties of a scalar field are defined by its couplings and
113: here we consider interactions with fermions as well as gauge
114: bosons. Within renormalizable models the former are given
115: by the Yukawa couplings
116: %---------- YR only
117: %of Eq.~(\ref{}). In our numerical analysis we
118: %consider top quark couplings of SM strength, i.e. $\chi_t = g_t/g_t^{SM} =1$
119: %for a scalar field, $H$, and $\chi_t = i$ for a pseudoscalar, $A$.
120: %--------- end YR
121: \begin{equation}
122: {\cal L}_Y = {y}_{f}\bar{\psi}H\psi +
123: \tilde{y}_{f}\bar{\psi}Ai\gamma_5\psi \;,
124: \label{eq:Yukawa}
125: \end{equation}
126: where $H$ (and $A$) denote (pseudo)scalar fields which couple to
127: fermions $\psi=t,\;b, \;\tau$ etc. In our numerical analysis we
128: consider couplings of SM strength, $y_f=\tilde y_f = m_f/v = y_{SM}$.
129: Via these Yukawa couplings, quark loops induce
130: effective couplings of the (pseudo)scalar to gluons which, for
131: (pseudo)scalar masses well below quark pair production threshold, can be
132: described by the effective Lagrangian
133: %--------- YR only
134: %\begin{equation}
135: %{\cal L}_{\rm eff} = {\rm Re}\chi_t \cdot
136: %\frac{\alpha_s}{12\pi v} \cdot H \, G_{\mu\nu}^a\,G^{a\,\mu\nu} +
137: %{\rm Im}\chi_t \cdot
138: %\frac{\alpha_s}{16\pi v} \cdot A \,
139: %G^{a}_{\mu\nu}\,G^{a}_{\rho\sigma}\varepsilon^{\mu\nu\rho\sigma}\;.
140: %\label{eq:ggS}
141: %\end{equation}
142: %-------- end YR only
143: \begin{equation}
144: {\cal L}_{\rm eff} = \frac{y_f}{y_{SM}}\cdot
145: \frac{\alpha_s}{12\pi v} \cdot H \, G_{\mu\nu}^a\,G^{a\,\mu\nu} +
146: \frac{\tilde y_f}{y_{SM}} \cdot
147: \frac{\alpha_s}{16\pi v} \cdot A \,
148: G^{a}_{\mu\nu}\,G^{a}_{\rho\sigma}\varepsilon^{\mu\nu\rho\sigma}\;.
149: \label{eq:ggS}
150: \end{equation}
151: Similar to the $\Phi gg$ coupling, Higgs couplings to $W$ and $Z$ bosons
152: will also receive contributions from heavy particle loops which can be
153: parameterized by the effective Lagrangian
154: \begin{equation}
155: {\cal L}_5 =
156: \frac{f_{\rm e}}{\Lambda_{5}} \; H \;
157: \vec{W}_{\mu\nu} \vec{W}^{\mu\nu}
158: + \frac{f_{\rm o}}{\Lambda_{5}} \; A \;
159: \vec{W}_{\mu\nu}
160: \vec{W}_{\rho\sigma}\,\frac{1}{2}\varepsilon^{\mu\nu\rho\sigma} \;.
161: \label{eq:d5}
162: \end{equation}
163:
164: For most models, one expects a coupling strength of order $f_i/\Lambda_5 \sim
165: \alpha/(4\pi v)$ for these dimension 5 couplings and, hence, cross
166: section contributions to vector boson fusion processes which are
167: suppressed by factors $\alpha/\pi$ (for interference effects with SM
168: contributions) or $(\alpha/\pi)^2$ compared to those mediated by the
169: tree level $HVV\; (V=W,\;Z)$ couplings of the SM. However, together with
170: the tree level
171: couplings, the effective Lagrangian of Eq.~(\ref{eq:d5}) has the virtue
172: that it parameterizes the most general $\Phi VV$ coupling which can
173: contribute in the vector boson fusion process $qQ\to qQ\Phi$ and, thus,
174: it is a convenient tool for phenomenological discussions and for
175: quantifying, to what extent certain couplings can be excluded experimentally.
176: Neglecting terms which vanish upon contraction with the conserved quark
177: currents, the most general tensor structure
178: for the fusion vertex $V^\mu(q_1)V^\nu(q_2)\to \Phi$ is given by
179: \begin{equation}
180: T^{\mu\nu}(q_1,q_2) = a_1(q_1,q_2)\;g^{\mu\nu} +
181: a_2(q_1,q_2)\;\left[ q_1\cdot q_2 g^{\mu\nu} - q_2^\mu q_1^\nu \right] +
182: a_3(q_1,q_2)\;\ \varepsilon^{\mu\nu\alpha\beta} q_{1\alpha}q_{2\beta} \;.
183: \end{equation}
184: Here the $a_i(q_1,q_2)$ are scalar form factors, which, in the low
185: energy limit, are given by the effective
186: Lagrangian of Eq.~(\ref{eq:d5}). One obtains, e.g. for the $W^+W^-\Phi$
187: coupling, $a_2 = -2f_e/\Lambda_5$ and $a_3 = 2f_o/\Lambda_5$, while
188: $a_1 = 2m_W^2/v$ is the SM vertex.
189:
190: \begin{figure}[htb]
191: \centerline{
192: \includegraphics[height=7.2cm, width=7.8cm]{squared_ww_bw.eps} \hspace*{0.3cm}
193: \includegraphics[scale=0.4]{gf_phijj.eps}
194: }
195: \caption[]{\label{fig:phijj} {\it Left:} Normalized distributions of the
196: azimuthal angle between the two
197: tagging jets, for the $\Phi\to WW\to e\mu\fdag{p}_T$ signal in vector boson
198: fusion at $m_\Phi = 160$~GeV, from Ref.~\protect\cite{Plehn:2001nj}.
199: Curves are for the SM and for single D5 operators as given in
200: Eq.~(\ref{eq:d5}), after cuts as in
201: Ref.~\protect\cite{Rainwater:1999sd}.
202: {\it Right:} The same for Higgs production in gluon fusion at
203: $m_\Phi=120$~GeV. Curves are for CP-even and CP-odd $\Phi tt$ coupling.
204: }
205:
206: \end{figure}
207:
208: The CP-even and CP-odd couplings of Eqs.~(\ref{eq:ggS},\ref{eq:d5}) lead
209: to characteristic azimuthal angle correlations of the two jets in $\Phi jj$
210: production processes. Normalized distributions of the azimuthal angle
211: between the two jets, $\triangle \phi_{jj}$, are shown in
212: Fig.~\ref{fig:phijj} for vector boson fusion (left panel) and for gluon
213: fusion processes (right panel) leading to $\Phi jj$ events: A CP-odd
214: coupling suppresses the cross section for planar events because
215: the epsilon tensor contracted with the four linearly dependent momentum
216: vectors of the incoming and outgoing partons disappears. For a CP-even
217: coupling the dip, instead, appears at 90
218: degrees~\cite{Plehn:2001nj,DelDuca:2001fn}. Unfortunately, when
219: both CP-even and CP-odd couplings are present simultaneously, the two
220: $\triangle \phi_{jj}$ distributions simply add, i.e. one does not
221: observe interference effects. The dip-structure which is present for pure
222: couplings is, thus, washed out.
223:
224: \begin{figure}[bht]
225:
226: \centerline{
227: \includegraphics[height=7.2cm, width=7.5cm]{azimuthal.eps} \hspace*{0.3cm}
228: \includegraphics[height=7.2cm, width=7.8cm]{pTdistr.eps}
229: }
230:
231: \caption[]{\label{fig:wbf} Normalized distributions of the tagging jets
232: in EW vector boson fusion with anomalous couplings and for a Higgs
233: mass of $m_\Phi =120$~GeV. Typical VBF cuts of $ p_{Tj} > 30\, {\rm GeV},\
234: |\eta_j| < 4.5,\ |\eta_{j_1}-\eta_{j_2}| > 4.0,\ m_{jj} > 600~{\rm GeV}$
235: are applied.
236: {\it Left:} Azimuthal angle distribution between the
237: two tagging jets, for different strengths of the operators of
238: Eq.~(\ref{eq:d5}).
239: {\it Right:} Transverse momentum distribution of the
240: hardest tagging jet for $f_e = f_o = 1$ and a form factor as in
241: Eq.~(\ref{eqn:C0}). The ``no formfactor'' curve corresponds to the
242: limit $M\to\infty$, i.e. a constant $a_i$.
243: }
244: \end{figure}
245:
246: This behavior is demonstrated in Fig.~\ref{fig:wbf}. For CP-even
247: and CP-odd couplings of the same strength, i.e. $f_e = f_o$, the
248: azimuthal angle distribution is very similar to the SM case. However,
249: in order to test the presence of anomalous couplings in such cases,
250: other jet distributions can be used, e.g. transverse
251: momentum distributions. The $\triangle \phi_{jj}$ distribution
252: is quite insensitive to variations of form factors, NLO corrections
253: and the like~\cite{Figy:2004pt}. On the other hand, $p_T$
254: distributions depend strongly on form factor effects. We study these
255: effects for a particular parameterization of the momentum dependence:
256: %
257: \begin{equation}
258: a_2(q_1,q_2) = a_3(q_1,q_2) \sim M^2 \; C_0 \ ( \ q_1, q_2, M \ )\;,
259: \label{eqn:C0}
260: \end{equation}
261: %
262: where $C_0$ is the familiar Passarino-Veltman scalar three-point
263: function~\cite{Passarino:1978jh}.
264: This ansatz is motivated by the fact that the $C_0$ function naturally
265: appears in the calculation of one-loop triangle diagrams, where the mass
266: scale M is given by the mass of the heavy particle in the loop.
267: As can be seen in the right panel of Fig.~\ref{fig:wbf}, even for a mass
268: scale M of the order of 50 GeV the anomalous couplings produce a harder
269: $p_T$ distribution of the tagging jets than is expected for SM
270: couplings. Thus it is possible to experimentally distinguish EW vector
271: boson fusion as predicted in the SM from loop induced $WW\Phi$ or
272: $ZZ\Phi$ couplings by the shape analysis of distributions alone.
273:
274: Let us now consider the gluon fusion processes where, for $\Phi tt$
275: couplings of SM strength, one does expect observable event rates from the
276: loop induced effective $\Phi gg$ couplings~\cite{DelDuca:2001fn}.
277: In order to assess the visibility of the CP-even vs. CP-odd signatures
278: of the azimuthal jet correlations at the LHC, we consider
279: Higgs + 2 jet production with
280: the Higgs decaying into a pair of $W$-bosons which further decay
281: leptonically, $\Phi \rightarrow W^+W^-\rightarrow \ell^+\ell^-\nu\bar{\nu}$. We
282: only consider electrons and muons ($\ell=e^\pm,\,\mu^\pm$) in the final
283: state. The Higgs-mass is set to $m_\Phi = 160$~GeV.
284: From previous studies on Higgs production in
285: vector boson fusion~\cite{Rainwater:1999sd} the main backgrounds are
286: known to be top-pair production i.e. $pp\rightarrow
287: t\bar{t},\,t\bar{t}j,\,t\bar{t}jj$~\cite{Kauer:2002sn}. The three cases
288: distinguish the number of $b$ quarks which emerge as tagging jets. The
289: $t\bar{t}$ case corresponds to both bottom-quarks from the
290: top-decays being identified as forward tagging jets, for $t\bar{t}j$
291: production only one tagging jet arises from a $b$ quark, while the $t\bar{t}jj$
292: cross section corresponds to both tagging jets arising from massless
293: partons. Further backgrounds arise from QCD induced $W^+W^-$ + 2
294: jet production and electroweak $W^+W^-jj$ production. These backgrounds are
295: calculated as in Refs.~\cite{Barger:1989yd} and~\cite{Jager:2006zc},
296: respectively. In the EW $W^+W^-jj$ background,
297: Higgs production in VBF is included, i.e. the VBF Higgs signal is
298: considered as a background to
299: the observation of $\Phi jj$ production in gluon fusion. We do not
300: consider backgrounds from $Zjj,\; Z\to\tau\tau$ and from $b\bar bjj$
301: production because they have been shown to be small in the
302: analyses of Refs.~\cite{Rainwater:1999sd,Kauer:2000hi}.
303:
304: \begin{table}[htb]
305: \caption{Signal rates and background cross sections for $m_\Phi=160\,{\rm
306: GeV}$. Results are given for the inclusive cuts of Eq.~(\ref{eqn:incl}),
307: the additional selection cuts of Eq.~(\ref{eqn:sel}) and b-quark
308: identification as discussed in the text, and with the additional
309: $\Delta\eta_{jj}$ cut of Eq.~(\ref{eqn:eta}) which improves the
310: sensitivity to the CP nature of the $\Phi tt$ coupling. The events
311: columns give the expected number of events for
312: ${ \cal L}_{int}=30\,{\rm fb}^{-1}$. }
313: \begin{center}
314: \begin{tabular}{|c|c|c|c|c|c|}
315: \cline{2-6}
316: \multicolumn{1}{c|}{} & inclusive cuts & \multicolumn{2}{c|}{selection
317: cuts} & \multicolumn{2}{c|}{selection
318: cuts + Eq.~(\ref{eqn:eta}) }\\
319: \hline
320: process & $\sigma$ [fb] & $\sigma$ [fb] & events / {$30\,{\rm fb}^{-1}$}
321: & $\sigma$ [fb] & events / {$30\,{\rm fb}^{-1}$}\\
322: \hline\hline
323: GF $pp \rightarrow \Phi +j j$ & 121.2 & {39.2} &{1176} & 13.1 & 393\\
324: VBF $pp \rightarrow W^+W^- +j j$ & 75.2 & {20.8} & {624} & 17.4 & 521\\
325: $pp \rightarrow t \bar{t}$ & 6832 & {29.6} & {888} & 2.0 & 60 \\
326: $pp \rightarrow t \bar{t}+ j$ & 9712 & {56.4} & {1692} & 15.6 & 468\\
327: $pp \rightarrow t \bar{t}+ jj $ & 1200 & {8.8} & {264} & 3.2 & 97\\
328: QCD $pp \rightarrow W^+W^- +j j$ & 364 & {15.2} & {456} & 3.9 & 116\\
329: \hline
330: \end{tabular}
331: \end{center}
332: \label{tab:xs}
333: \end{table}
334:
335: \begin{figure}[htb]
336: \centerline{
337: \includegraphics[scale=0.4]{azimuth_even3.eps}
338: \includegraphics[scale=0.4]{azimuth_odd3.eps}
339: }
340: \caption[]{\label{fig:events} Distribution of the azimuthal angle
341: between the tagging jets in $\Phi jj$ events
342: for a CP-even $\Phi tt$ coupling ({\it left}) and a CP-odd coupling
343: ({\it right}). Shown are expected signal and background events per
344: 10 degree bin for
345: $\Phi \rightarrow W^+W^-\rightarrow \ell^+\ell^-\nu\bar{\nu}$ and
346: ${\cal L}_{int}=30\,{\rm fb}^{-1}$ for
347: the cuts of Eqs. (\ref{eqn:incl}, \ref{eqn:sel}, \ref{eqn:eta}) and an applied b-veto.
348: Processes from top to bottom: gluon fusion (signal), VBF, $t\bar{t}$,
349: $t\bar{t}j$, $t\bar{t}jj$, QCD $WWjj$. $m_{\Phi}=160$~GeV is assumed.
350: }
351: \end{figure}
352:
353:
354: The inclusive cuts in Eq.~(\ref{eqn:incl}) reflect the requirement
355: that the two tagging jets and two charged leptons are observed inside
356: the detector, and are well-separated from each other.
357: \begin{equation*}
358: p_{Tj} > 30\, {\rm GeV},\qquad|\eta_j| < 4.5,\qquad |\eta_{j_1}-\eta_{j_2}| > 1.0
359: \end{equation*}
360: \begin{equation}
361: p_{T\ell} > 10\, {\rm GeV},\qquad|\eta_\ell| < 2.5,\qquad \Delta R_{j\ell} > 0.7
362: \label{eqn:incl}
363: \end{equation}
364: The resulting cross sections for these cuts are shown in Table
365: \ref{tab:xs}. The signal cross section of 121~fb (which includes the
366: branching ratios into leptons) is quite sizeable.
367: The QCD $WWjj$ cross section is about 3 times higher whereas the VBF
368: process reaches $2/3$ of the signal rate. The worst source of background
369: arises from the $t\bar{t}$ processes, with a total cross section of more than
370: 17~pb.\\
371: In order to improve the signal to background ratio the following
372: selection cuts are applied:
373: \begin{equation*}
374: p_{T\ell} > 30\, {\rm GeV},\qquad m_{\ell\ell} < 75\, {\rm GeV},\qquad
375: \Delta R_{\ell\ell} < 1.1
376: \end{equation*}
377: \begin{equation}
378: m_{T}^{WW} < 170\, {\rm GeV},\qquad m_{\ell\ell} < 0.5\cdot m_{T}^{WW}\;.
379: \label{eqn:sel}
380: \end{equation}
381: Here, the transverse mass of the dilepton-$\fdag{\vec{p}}_T$ system is
382: defined as~\cite{Rainwater:1999sd}
383: \begin{equation}
384: m_T^{WW} = \sqrt{(\fdag{E}_T+E_{T,{\ell\ell}})^2-
385: ({\vec{p}}_{T,{\ell\ell}}+\fdag{\vec{p}}_T)^2}
386: \end{equation}
387: in terms of the invariant mass of the two charged lepton and the
388: transverse energies
389: \begin{equation}
390: E_{T,\ell\ell} = (p_{T,\ell\ell}^2 + m_{\ell\ell}^2)^{1/2},\qquad
391: \fdag{E}_T = (\fdag{p}_T^2 + m_{\ell\ell}^2)^{1/2} .
392: \end{equation}
393: %{\it ...Explanation of the cuts...}\\
394: In addition to these cuts we make use of a b-veto to reduce the
395: large top-background. We reject all events where at least one jet is identified
396: as a b-jet. Using numbers from Ref.~\cite{Weiser:2006md}, we assume b-tagging
397: efficiencies in the range of $60\%-75\%$ (depending on b-rapidity and
398: transverse momentum) and an overall mistagging probability of $10\%$ for
399: light partons.\\
400: With the selection cuts (\ref{eqn:sel}) and the b-veto the backgrounds
401: can be strongly suppressed. Table \ref{tab:xs} shows the resulting
402: cross sections and the expected number of events for an integrated
403: luminosity of ${\cal L}_{int}=30\,{\rm fb}^{-1}$. The signal rate is reduced
404: by a factor of 3 but the backgrounds now have cross sections of the same
405: order as the signal. The largest background still comes from the
406: $t\bar{t}$ processes,
407: especially $t\bar{t}+1j$. For $30\,{\rm fb}^{-1}$ we get about $1000$
408: signal events on top of $4000$ background events. This corresponds to a
409: purely statistical significance of the gluon fusion signal of
410: $S/\sqrt{B}\approx 18$ and a sufficient number of events
411: to analyze the azimuthal jet correlations.
412:
413: Fig.~\ref{fig:events} shows the expected $\triangle \phi_{jj}$
414: distribution for $30\,{\rm fb}^{-1}$. Plotted are
415: signal events on top of the various backgrounds.
416: An additional cut on the rapidity gap between the jets
417: \begin{equation}
418: |\eta_{j_1}-\eta_{j_2}| > 3.0
419: \label{eqn:eta}
420: \end{equation}
421: has been applied. It enhances the shape of the distribution that is
422: sensitive to the nature of the $\Phi tt$ coupling. Clearly visible, the
423: distribution for the CP-even coupling has a slight minimum at $\triangle
424: \phi_{jj} = 90^\circ$ whereas for the CP-odd case there is a pronounced
425: maximum. In order to quantify this, we define the fit-function
426: \begin{equation}
427: f(\triangle \phi) = C\cdot(1+A\cdot\cos{2\triangle
428: \phi}+B\cdot\cos{\triangle \phi})
429: \end{equation}
430: with free parameters $A$, $B$, $C$.
431: The fit is shown as black curves in Fig. \ref{fig:events}. The parameter
432: $A$ is now a measure for the $\triangle \phi_{jj}$ asymmetry,
433: i.e. whether there is a CP-even or CP-odd $\Phi tt$ coupling. The fitted
434: values are $A = 0.064 \pm 0.035$ for the CP-even and $A=-0.157 \pm
435: 0.034$ for the CP-odd case, while $A_{B} = -0.039 \pm 0.040$ for the
436: sum of all backgrounds. Defining a significance $s$ as
437: \begin{equation}
438: s = \frac{(A_{S+B}-A_{B})}{\Delta A_{S+B}}\; ,
439: \end{equation}
440: we get $s = 3.0$ and $s = -3.4$ for the CP-even and CP-odd case,
441: respectively.
442: Thus, a distinction of a CP-odd and CP-even $\Phi tt$ coupling is possible
443: at a $6\sigma$ level for the considered process and a Higgs mass of
444: 160~GeV. This implies that, at least for favorable values of the Higgs
445: boson mass, (i) an effective separation of VBF and gluon fusion sources
446: of $\Phi jj$ events is possible and (ii) the CP nature of the $\Phi tt$
447: coupling of Eq.~(\ref{eq:Yukawa}) can be determined at the LHC.
448:
449:
450:
451: %---------------------------------------------------------------------
452: \section*{Acknowledgments}
453: %---------------------------------------------------------------------
454:
455: This research was supported by the Deutsche
456: Forschungsgemeinschaft in the Sonderforschungsbereich/Transregio
457: SFB/TR-9 ``Computational Particle Physics'' and in the Graduiertenkolleg
458: "High Energy Physics and Particle Astrophysics".
459:
460: \bibliography{hxjj}
461:
462: \end{document}
463:
464: