Abstract
TauDecay is a library of helicity amplitudes to simulate polarized tau decays, constructed in the FeynRules and MadGraph5 framework. Together with the leptonic mode, the decay library includes the main hadronic modes, τ→ν _{ τ }+π, 2π, and 3π, which are introduced as effective vertices by using FeynRules. The model file allows us to simulate tau decays when the onshell tau production is kinematically forbidden. We also demonstrate that all possible correlations among the decay products of pairproduced taus through a Z boson and a scalar/pseudoscalar Higgs boson are produced automatically. The program has been tested carefully by making use of the standard tau decay library Tauola.
1 Introduction
The recent observation of a standardmodellike Higgs boson with mass around 125 GeV at the LHC [1, 2] has increased our interests in the H→τ ^{+} τ ^{−} decay mode, which has the modest branching fraction yet to be established [3, 4]. On the other hand, tau leptons have been well known as a particle which can explore the Higgs sector [5–8] as well as supersymmetric models [9–11], through their polarization [12–14].
Tauola [15–17] is a wellknown program to simulate polarized τ decays, including various hadronic modes, and recent theoretical and experimental event simulations involving τ decays heavily depend on it.^{Footnote 1} The standard Tauola Universal Interface [22, 23] includes the longitudinal spin correlations between pairproduced taus [24], while the extended version also takes into account the transverse spin effects [25–28]. We point here some limitations of the use of Tauola:

(i)
The produced τ has to be stable, i.e. onshell, in event generators.

(ii)
The interface for the spin correlation effects is limited to the standard processes such as Z/γ ^{∗}→τ ^{+} τ ^{−} and H/A→τ ^{+} τ ^{−}.

(iii)
For the transverse spin effects, a particular simulation is required by accepting or rejecting a pair of produced taus based on a spin weighting factor.
In this article, we present a new implementation of a τdecay model file, and construct a library of helicity amplitudes, TauDecay,^{Footnote 2} to simulate polarized τ decays in the FeynRules (FR) [29] and MadGraph5 (MG5) [30] framework. In Sect. 2 we describe how we implement the effective vertices via FR for the main hadronic decay modes, namely τ→ν _{ τ }+π, 2π, and 3π, and present the validation of our TauDecay package. To address the limitations mentioned above, in Sect. 3 we study scalartau (stau) decays in scenarios with stau being nearly degenerate in mass with neutralino, where the onshell tau production is kinematically forbidden. Moreover, in Sects. 4 and 5 we demonstrate that all possible correlations among the decay products of pairproduced taus via a Z boson and a scalar/pseudoscalar Higgs boson can be produced within our fullfledged package. Section 6 presents our brief summary.
2 Effective vertices for hadronic tau decays
In this section, we consider effective vertices for the three hadronic τ decay modes, which together with the leptonic mode account for 90 % of the τ decays [31], and describe how we implemented them via the FeynRules (FR) [29] and MadGraph5 (MG5) [30] packages.
The decay modes which we consider in this article are
as the 2π and 3π modes are dominated by the ρ and a _{1} vectormeson productions, respectively.
The τ–ν _{ τ }–π vertex can be introduced by the effective interaction Lagrangian:
with the physical constants in Table 1, the chiral projection operator P _{ L }, and the constant form factor
The effective Lagrangian for the ρ mode is
with
The form factor F _{ ρ }(Q ^{2}) is parametrized by [15, 32]
with the Breit–Wigner factor
where Q=q _{1}+q _{2} for τ ^{−}→νπ ^{−}(q _{1})π ^{0}(q _{2}). The running width is
where the ρ meson line shape factor is
with
In practice, we introduce only pions, π ^{±} and π ^{0}, as new particles and implement the above Lagrangians (2) and (4) into FR,^{Footnote 3} which provides the Ufo (Universal FeynRules Output) model file [33], with f _{2} as a constant parameter. The Aloha (Automatic Libraries Of Helicity Amplitudes) program [34] in MG5 reads the model file to create the Helas (HELicity Amplitude Subroutines) library [35, 36] for helicity amplitude computations. At this stage we replace the constant parameter by the momentumdependent form factor (5).
Unlike the above two cases, the effective vertex for the a _{1} mode cannot be obtained by the Lagrangian since the vertex structure is not symmetric between the two identical pion in the final state. Therefore, instead of introducing the Lagrangian, we implement it by hand in FR and created the Ufo model file.^{Footnote 4} The effective vertex, or the decay amplitude, for the a _{1} mode is
where the hadronic current J ^{μ} is given by [15, 32]
with Q=q _{1}+q _{2}+q _{3} for τ ^{−}→ν _{ τ } π ^{−}(q _{1})π ^{−}(q _{2})π ^{+}(q _{3}). The form factors are
with Q _{ i3}=q _{ i }+q _{3} (i=1,2). The a _{1} meson line shape factor in (8) is parametrized as [15, 32]
The π ^{0} π ^{0} π ^{−} mode is the same, except the mass difference between π ^{±} and π ^{0}. The relevant particle masses and widths are collected in Table 2.
Our Ufo model now allows MG5 to generate Feynman diagrams including hadronic τ decays (Fig. 1) and the corresponding helicity amplitudes, while the leptonic decay mode can be simulated within the default Standard Model Ufo via the offshell W boson. TauDecay is a library to simulate polarized τ decays, based on the τ decay helicity amplitudes created by MG5 with the form factor implementation. Using TauDecay, we produce the partial decay widths in Table 3 as well as the pion invariant mass distributions in Fig. 2 in good agreement with the standard τ decay library Tauola [17].^{Footnote 5} We note that the slight mismatch between the two programs in the table as well as in the large \(m_{\pi^{}\pi^{0}}\) invariant mass region for the ρ mode comes from the QED correction for the leptonic mode in Tauola and the different parameter choice, e.g. f _{ π } and m _{ ρ′}. Moreover, the fractional energy distributions of the polarized τ decays in Fig. 3 agree with the ones in the collinear limit [5–7] as well as by Tauola.
3 Stau decays via an offshell tau
An important application of our FR τdecay model file is for the case that onshell τ production is kinematically forbidden. The model file allows MG5 to treat an intermediate τ as a propagator.
Such a case can happen in the socalled stauneutralino (\(\tilde{\tau}_{1}\)\(\tilde{\chi}^{0}_{1}\)) coannihilation scenario in the constrained minimal supersymmetric standard model (CMSSM). The neutralino \(\tilde{\chi}^{0}_{1}\) is the lightest supersymmetric particle (LSP) and stable by assuming the Rparity conservation. The scalartau (stau) \(\tilde{\tau}_{1}\) is the nexttolightest SUSY particle (NLSP), which exclusively decays into a τlepton and a LSP \(\tilde{\chi}^{0}_{1}\). The scenario is cosmologically preferred to provide the observed dark matter relic density [38] as well as to solve the ^{7}Li problem in the standard bigbang nucleosynthesis [39]. The recent global fit analysis for the CMSSM also favors the point [40]. If \(m_{\tilde{\tau}_{1}}m_{\tilde{\chi}^{0}_{1}}>m_{\tau}\), there is no obstacle to use Tauola for τ decays, following event generation with τleptons as an onshell particle [41]. For the case of
however, we cannot generate events unless τ decays are taken into account in event generators. One of the common benchmark points for the SUSY searches at the LHC is e.g. CMSSM40.1.2 [42], which provides the above mass spectrum with \(m_{\tilde{\tau}_{1}}=230.3\) GeV and \(m_{\tilde{\chi}^{0}_{1}}=228.7\) GeV [43].^{Footnote 6}
When the twobody decay mode \(\tilde{\tau}_{1}\to\tilde{\chi}^{0}_{1}+\tau\) is closed, the threebody or fourbody decays via the offshell τ (see, e.g. Fig. 4) become dominant and the \(\tilde{\tau}_{1}\) could be longlived [40, 46, 47]. Figure 5 shows the stau partial decay widths for each τ decay mode against the \(\tilde{\chi}^{0}_{1}\) mass. The \(\tilde{\tau}_{1}\) mass and the relevant mixing angles in the neutralino and stau sectors are fixed as in CMSSM40.1.2, providing a \(\tilde{\tau}_{R}\)like stau and a binolike neutralino. The vertical dashed lines denote the threshold for each decay mode, where a _{1} and ρ indicate Δm=m _{3π } and m _{2π }, respectively. As the \(\tilde{\chi}^{0}_{1}\) mass is approaching the \(\tilde{\tau}_{1}\) mass, i.e. Δm becomes smaller, the decay width becomes smaller very rapidly. Just below the τ threshold, around the CMSSM40.1.2 point (denoted by a vertical solid line), decay widths of all the modes are comparable with those in the onshell τ case, while the π mode is dominant in the small Δm region due to the phase space suppression for the multipion and leptonic modes. Only in the Δm<m _{ π } region the electronic mode becomes dominant. Those agree with the recent study [40], although ρ and a _{1} are considered as onshell particles in their calculations.
The collider signature of the longlived \(\tilde{\tau}_{1}\) significantly depends on the \(\tilde{\tau}_{1}\) width, i.e. the lifetime; see e.g. Refs. [48–50], and as shown in Fig. 5 the lifetime strongly depends on Δm. For instance, the \(\tilde{\tau}_{1}\) predicted at CMSSM40.1.2, whose lifetime is \(\mathcal{O}(10^{8}~\mathrm{s})\), could leave a charged track with a displaced vertex of its decay inside the detectors. We briefly study the decay distributions for such a longlived stau to see the leftright mixing of \(\tilde{\tau}_{1}\) in this situation.
The scalar partners of lefthanded and righthanded taus, \(\tilde{\tau}_{L}\) and \(\tilde{\tau}_{R}\), mix to form two mass eigenstates, and the lighter one is
where \(\theta_{\tilde{\tau}}\) is the mixing angle. Similarly gauginos, \(\tilde{B}\) and \(\tilde{W}_{3}\), and neutral higgsinos, \(\tilde{H}_{d}^{0}\) and \(\tilde{H}_{u}^{0}\), mix to form four mass eigenstates, and the lightest one is
where \(\tilde{X}=(\tilde{B},\tilde{W}_{3},\tilde{H}_{d}^{0},\tilde{H}_{u}^{0})\). The chirality of τ in the \(\tilde{\tau}_{1}\to\tau\tilde{\chi}^{0}_{1}\) decay depends on these mixings and determined by the interaction Lagrangian
with the chiralprojection operators, \(P_{R/L}=\frac{1}{2}(1\pm\gamma^{5})\). The couplings, a _{ L } and a _{ R }, are
where \(Y_{\tau}=gm_{\tau}/\sqrt{2}m_{W}\cos\beta\), g is the SU(2)_{ L } gauge coupling, and tanβ is the ratio of the vacuum expectation values of the two Higgs doublets. For most of SUSY scenarios as well as for CMSSM40.1.2 the lightest neutralino is gauginolike (U _{31}∼0), where the chirality is conserved.
Figure 6 shows the energy distributions of the oneprong pion in the π mode (left) and in the ρ mode (right) for the lefthanded \(\tilde{\tau}\) (blue) and righthanded \(\tilde{\tau}\) (red) decay in the stau rest frame. The distributions for the reconstructed ρ is also shown by dashed lines as a reference. The phase space of the decaying offshell τ is limited and it is no longer energetic. Therefore the energy distributions of the pion are quite different from the ones in Fig. 3, where the collinear limit is safely applied, although the remnant of the difference between lefthanded and righthanded can be still seen especially for the π and ρ modes.
4 Spin correlations in tau pair decays
Another important application of the FR τdecay model is spin correlations in taupair production [5–8, 18–21, 24–28].^{Footnote 7} As in the previous section, one can generate amplitudes including the τ decays, and hence the spin correlations between the two taus are automatically taken into account. In other words, there is no need for a particular simulation to get the transverse spin effects.
In this section, we present correlations in taupair production through a Z boson, a CPeven scalar H, or a CPodd scalar A,
and the subsequent hadronic τ decays. We note that, although we consider the above standard τpair production processes in this paper, a special interface to cooperate with any new physics models is not required when we generate amplitudes with intermediate taus being as propagators.
4.1 Helicity formalism
First, we discuss the case that both taus decay to ν _{ τ } π by using analytic expressions. We assign the fourmomentum and the helicity of each particle for the process as
where a _{1,2} stand for quarks or gluons (or even electrons or photons in case of e ^{+} e ^{−} or γγ collisions). The helicities take the values σ _{ i }/2 and λ _{ i }/2 for quarks and leptons and σ _{ i } for gluons and photons with σ _{ i }=±1 and λ _{ i }=±1. Although we evaluate the full amplitude for the above processes numerically to present the distributions, we discuss it below in terms of the τpair production amplitude and the τ decay amplitudes, which give us better understanding of the distributions. Moreover, we construct our TauDecay package based on the following analytic expressions in the end of this section.
Using the completeness relations
the full amplitude can be expressed as the product of the taupair production amplitude (\(\mathcal{M}_{X=Z,H,A}\)) and two τ→πν decay amplitudes (\(\mathcal{M}_{1,2}\)):
with the τ propagator factor D(q ^{2})=(q ^{2}−m ^{2}+imΓ)^{−1}.
It is straightforward to obtain the squared matrix elements of the full production plus decay amplitudes,
in terms of the production density matrix \({\mathcal{P}}^{\lambda_{1}\lambda_{2}}_{\bar{\lambda}_{1}\bar{\lambda}_{2}}\) and the decay density matrices \({\mathcal{D}_{1,2}}^{\lambda_{1,2}}_{\bar{\lambda}_{1,2}}\);
In the narrow width limit the propagator factor becomes
Although we only consider partonlevel subprocesses, one can generalize (25) to mixed case and apply it for any processes, including summation over different subprocesses and a product of the relevant parton densities. We can also easily replace the τ→ν _{ τ } π decay amplitude by one for the other hadronic decay mode as well as the leptonic mode.
4.2 Kinematics
Let us define the kinematical variables. In the collision centerofmass (CM) frame, i.e. in the X rest frame, we choose the τ momentum direction as the zaxis,
where \(\beta=\bar{\beta}(\frac{q_{1}^{2}}{\hat{s}},\frac{q_{2}^{2}}{\hat{s}} )\) with \(\bar{\beta}(a,b)\) defined in (10), Θ is the scattering angle, and we choose p _{1}×q _{1} direction as the yaxis. The momenta of the τ ^{−} decay products are parametrized in the τ ^{−} rest frame,
with \(\beta_{1}=1m_{\pi}^{2}/q_{1}^{2}\). Similarly, those of the τ ^{+} decay products are
with \(\beta_{2}=1m_{\pi}^{2}/q_{2}^{2}\). We note that the polar (z)axis and the yaxis normal to the scattering plane are chosen common to all the three frames, and the two decay frames differ only by the boost along the τ production axis (see Fig. 7). The τ width is very narrow, \(\varGamma\sim\mathcal{O}(10^{12}~\mathrm{GeV})\), and hence we take the narrow width limit, \(q_{1}^{2}=q_{2}^{2}=m^{2}\), in the following analytic amplitudes.
4.3 Taupair production and decay amplitudes
The helicity amplitude for the taupair production via Zboson (21a) is given by
where the reduced amplitude is
Here, we neglect the initial fermion mass and take σ=σ _{1}=−σ _{2}. \(g_{\pm}^{q}\) and \(g_{\pm}^{\tau}\) are the Zboson couplings to right and lefthanded quarks and tau leptons, respectively.
The amplitude for the S(=H,A)boson case (21b) and (21c) is
where
The amplitudes are nonzero only when σ=σ _{1}=σ _{2} and λ=λ _{1}=λ _{2}. Here, \(y_{\tau}=\sqrt{2}m_{\tau }/v\) is the τ Yukawa coupling, and g _{ S } is the ggS coupling, which is given by g _{ H }=α _{ s } g _{ Htt }/3πv and g _{ A }=α _{ s } g _{ Att }/2πv in the heavytop limit.
The τ→πν decay amplitudes in the τ rest frame are
where
Those amplitudes are invariant with the boost to the collision CM frame.
4.4 Helicity correlations
The trivial helicity correlations are given by the diagonal parts of the density matrices, \(\lambda_{1}=\bar{\lambda}_{1}\) and \(\lambda _{2}=\bar{\lambda}_{2}\). In the following analytic expressions we take \(m_{\tau}/\sqrt{\hat{s}}=0\), i.e. β=1.
For the Z production, only for λ _{1}≠λ _{2} the production amplitude is nonzero. After integrating out the scattering angle Θ and the azimuthal angles ϕ _{1,2}, the squared matrix element (25) is
with \(\kappa=({g_{+}^{\tau}}^{2}{g_{}^{\tau}}^{2})/({g_{+}^{\tau}}^{2}+{g_{}^{\tau }}^{2})\sim 0.15\). As seen in (39a), (39b), the polar angle cosθ _{1,2} distribution of the pion arising from \(\tau^{\pm}_{L}\) is (1−cosθ _{1,2}), while from \(\tau^{\pm}_{R}\) it is (1+cosθ _{1,2}). Hence in \(Z\to\tau^{+}_{L}\tau^{}_{R}\) or \(\tau ^{+}_{R}\tau^{}_{L}\) decays the two pions tend to be emitted to the opposite direction along the zaxis. The difference between the left and right coupling, i.e. parity violation, gives a small linear dependence of cosθ _{1} and cosθ _{2} in the single differential cross section, and one can see a slightly higher density in the cosθ _{1}<0 and cosθ _{2}>0 region in the left panel of Fig. 8.
For the Higgs production, on the other hand, only for λ _{1}=λ _{2} the production amplitude is nonzero, and therefore
Apart from the linear dependence of cosθ _{1,2} for the Z case, completely the opposite is favored for \(H/A\to\tau^{+}_{L}\tau^{}_{L}\) or \(\tau^{+}_{R}\tau^{}_{R}\) decays as shown in the right panel of Fig. 8. Note that the helicity correlations for H and A are identical. For comparison the masses of H and A are assumed to be the Zboson mass, m _{ H/A }=m _{ Z }.
4.5 Polarization correlations
The nontrivial polarization correlations are given by the offdiagonal parts of the density matrices, i.e. \(\lambda_{1}=\bar{\lambda}_{1}\) and \(\lambda_{2}=\bar{\lambda}_{2}\), which produce the azimuthal angle dependence. When we isolate the azimuthal angle dependence in (25), there are nine distributions (including one constant piece) as
Here, and in the following, summation over repeated indices \((\lambda_{1},\lambda_{2},\bar{\lambda}_{1},\bar{\lambda}_{2})=\pm\) is implied. The coefficients \(F_{i}^{(\pm)}\) are the functions of the kinematical variables except the azimuthal angles ϕ _{1,2}. For the production of a Z boson, they also depend on the production angle Θ. For the spin0 particle case, only the \(F_{4}^{}\) term in (42) survives due to the helicity selection λ _{1}=λ _{2}. All the sine terms vanish when CP is conserved and when the absorptive part of the amplitudes are neglected, e.g., in the treelevel approximation.
Because the phase of the product of the two decay density matrices is
the coefficients \(F_{14}^{(\pm)}\) are expressed in terms of the production density matrix and two decay density matrices as
The azimuthal angle correlations are manifestly expressed by quantum interference among different helicity states of the intermediate tau leptons.
The Z production could produce the ϕ _{1} (or ϕ _{2}) dependence, however this is very small because F _{2} (F _{3}) is proportional to \(m/\sqrt{s}\). Similarly, \(F_{4}^{}\propto\mathcal{O}(m^{2}/\hat{s})\), and hence the distribution of the azimuthal difference between the two τdecay planes, ϕ _{1}−ϕ _{2}(≡Δϕ), is flat. An interesting correlation between the two decay planes for the Z case is the ϕ _{1}+ϕ _{2}(≡Φ) correlation, whose coefficient \(F_{4}^{+}\) is
After integrating out Θ, θ _{1} and θ _{2}, the azimuthal asymmetry is given by
with
The distribution is enhanced around Φ=0 and 2π, while it is suppressed around Φ=π.
For the H/A production, as mentioned above, only the ϕ _{1}−ϕ _{2} term is nonzero and the coefficient is
where the −/+ sign is for CPeven/odd scalar. The azimuthal asymmetry is given by [8]
with
The CPeven and odd scalars have opposite modulation, and the distribution is strongly enhanced (suppressed) around Δϕ=π for the CPeven (odd) scalars.
To examine the validity of the model file, Fig. 9 shows the normalized azimuthal correlations between π ^{−} and π ^{+}, and our numerical results (solid histograms) agree well with the above analytic formula (dotted lines).
4.6 Spin correlations at the LHC
To present the validation of our program, we have so far discussed the angular distributions, assuming that all the τdecay products can be reconstructed. Here, we show some distributions of the realistic experimental observables at the LHC.
First, we extend the helicity correlation in τ→ν _{ τ } π discussed above to other tau decay modes. In Fig. 10 we present the helicity correlations in terms of energy fractions in the laboratory frame for \(pp\to Z/H/A\to\tau^{}(\to\pi^{}\nu)\tau^{+}(\to A^{+}\bar{\nu})\), where A ^{+} is the leptonic, π, ρ, and a _{1} decay modes, respectively, from left to right. The oneprong τ decays are considered and the energy fractions are \(z_{1}=E_{\pi^{}}/E_{\tau^{}}\) and \(z_{2}=E_{\hat{A}^{+}}/E_{\tau^{+}}\) with \(\hat{A}=e,\pi,\pi(\rho),\pi(a_{1})\). For comparison, the resonant mass is taken to the Zboson mass, and hence the both taus are highly boosted, where the collinear approximation can be safely applied. For the A=π mode, in the collinear limit (β=1) we have
As a result, the z _{1}z _{2} correlation behaves opposite to cosθ _{1}cosθ _{2} correlation as shown in Fig. 8. The helicity correlations are different between the spin1 and spin0 bosons, while those are identical between the CPeven and odd scalars.
Second, an useful observable to measure the parity of Higgs boson is the acollinearity angle, namely the angle δ between π ^{+} and π ^{−} [8]. The azimuthal asymmetry Δϕ can be given as a function of angle δ. To display the measurable difference of Higgs parities at the LHC, in Fig. 11, we show the π ^{+} π ^{−} acollinearity distribution in the Higgs rest frame (left) and in the lab frame (right) for \(pp\to H/A\to\tau^{}(\to\pi^{}\nu)\tau^{+}(\to\pi^{+}\bar{\nu})\) at \(\sqrt{s}=8\) TeV. Here we take the Higgs mass as 125 GeV. One can see the distinguishable difference between CPeven and CPodd scalars around δ=π, although in the lab frame the boost effect leads to relatively unclear discrepancy. While the reconstruction of the Higgs rest frame at hadron colliers is difficult, it is possible in the Zassociated Higgs production at e ^{+} e ^{−} colliders and the acollinearity distribution in the reconstructed H/A rest frame has been discussed in [25].
5 Spin correlations with TauDecay
To validate our FR τdecay model file, all the results shown in Sects. 3 and 4 are generated in MG5 with intermediate taus being as propagators, denoted by method A. Unless intermediate taus are offshell such as the case in Sect. 3, a library which collects all the possible τ decay channels would be much more efficient for practical event generations since τ production and its decay can be simulated independently as we use Tauola.
TauDecay is a Fortran library of τdecay helicity amplitudes, constructed in the FR/MG5 framework as described in Sect. 2. The input of the subroutines for each decay channel is the four momenta of the decaying τ as well as its decay products and their helicities. The output is the amplitude, i.e. a complex number, at a given phase space point with a given helicity configuration. In this section, we present how the TauDecay package can reproduce spin correlation discussed in Sect. 4.
We show two methods by means of TauDecay; without and with spin density matrix, denoted by method B and C, respectively. The both methods are carefully checked by method A.
 B::

(without density matrix) The τ production amplitude as well as the decay amplitudes from TauDecay are evaluated for a given process for a given phase space point. The τ helicities in the product of the production and decay amplitudes should be summed over as indicated in (24), and then the τ helicity summed amplitudes should be squared. This is theoretically identical to the propagator method in the narrow width limit, but it is useful for systematically studying different tau decay modes for each production process.
 C::

(with density matrix) First, the τ production density matrix is computed for each phase space point. If there is only one tau in the final state, this will be a 2×2 matrix, while it will be a 4×4 double density matrix for two taus as \({\mathcal{P}}^{\lambda_{1}\lambda_{2}}_{\bar{\lambda}_{1}\bar{\lambda}_{2}}\) in (26). Then, the correlated tau decays are simulated by multiplying the production density matrix with the corresponding τ decay helicity amplitudes and their complex conjugates provided by TauDecay, namely the decay density matrices \({\mathcal{D}_{i}}^{\lambda_{i}}_{\bar{\lambda}_{i}}\), according to (25).
Figure 12 shows the azimuthal angle Δϕ correlation in \(pp\to H/A\to\tau^{}(\to\pi^{}\nu)\tau^{+}(\to\pi^{+}\bar{\nu})\) for the comparison of the three methods, and the same results are reproduced as expected.
6 Summary
In this paper, we have implemented the main hadronic tau decays, τ→ν _{ τ }+π, 2π, and 3π, by using the effective vertices in the FeynRules and MadGraph5 framework, and constructed TauDecay, a library of helicity amplitudes to simulate polarized tau decays. The model file allows us to simulate tau decays when the onshell tau production is kinematically forbidden, and the stau decay in the stauneutralino coannihilation region was discussed as an application. We also demonstrated that all possible correlations among the decay products of pairproduced taus through a Zboson and a scalar/pseudoscalar Higgs boson can be produced within our fullfledged package. The program has been tested carefully by making use of the standard tau decay library Tauola.
Notes
The TauDecay package is available at http://madgraph.kek.jp/KEK/.
π ^{0} decay is not considered in the program.
Input option of the function WriteUFO in FR allows us to include such a vertex [33].
The correlation of τ ^{±} τ ^{±} from a doubly charged Higgs boson H ^{±±} has also been studied in [51].
References
G. Aad et al. (ATLAS Collaboration), Phys. Lett. B 716, 1 (2012). arXiv:1207.7214 [hepex]
S. Chatrchyan et al. (CMS Collaboration), Phys. Lett. B 716, 30 (2012). arXiv:1207.7235 [hepex]
S. Chatrchyan et al. (CMS Collaboration), Phys. Lett. B 713, 68 (2012). arXiv:1202.4083 [hepex]
G. Aad et al. (ATLAS Collaboration), J. High Energy Phys. 1209, 070 (2012). arXiv:1206.5971 [hepex]
B.K. Bullock, K. Hagiwara, A.D. Martin, Phys. Rev. Lett. 67, 3055 (1991)
B.K. Bullock, K. Hagiwara, A.D. Martin, Phys. Lett. B 273, 501 (1991)
B.K. Bullock, K. Hagiwara, A.D. Martin, Nucl. Phys. B 395, 499 (1993)
M. Kramer, J.H. Kuhn, M.L. Stong, P.M. Zerwas, Z. Phys. C 64, 21 (1994). hepph/9404280
M.M. Nojiri, Phys. Rev. D 51, 6281 (1995). hepph/9412374
M.M. Nojiri, K. Fujii, T. Tsukamoto, Phys. Rev. D 54, 6756 (1996). hepph/9606370
S.Y. Choi, K. Hagiwara, Y.G. Kim, K. Mawatari, P.M. Zerwas, Phys. Lett. B 648, 207 (2007). hepph/0612237
Y.S. Tsai, Phys. Rev. D 4, 2821 (1971) [Erratumibid. D 13, 771 (1976)]
K. Hagiwara, A.D. Martin, D. Zeppenfeld, Phys. Lett. B 235, 198 (1990)
A. Rouge, Z. Phys. C 48, 75 (1990)
S. Jadach, J.H. Kuhn, Z. Was, Comput. Phys. Commun. 64, 275 (1990)
M. Jezabek, Z. Was, S. Jadach, J.H. Kuhn, Comput. Phys. Commun. 70, 69 (1992)
S. Jadach, Z. Was, R. Decker, J.H. Kuhn, Comput. Phys. Commun. 76, 361 (1993)
D. Grellscheid, P. Richardson, arXiv:0710.1951 [hepph]
P. Ilten, arXiv:1211.6730 [hepph]
Z. Czyczula, T. Przedzinski, Z. Was, Eur. Phys. J. C 72, 1988 (2012). arXiv:1201.0117 [hepph]
S. Banerjee, J. Kalinowski, W. Kotlarski, T. Przedzinski, Z. Was. arXiv:1212.2873 [hepph]
P. Golonka, B. Kersevan, T. Pierzchala, E. RichterWas, Z. Was, M. Worek, Comput. Phys. Commun. 174, 818 (2006). hepph/0312240
N. Davidson, G. Nanava, T. Przedzinski, E. RichterWas, Z. Was, Comput. Phys. Commun. 183, 821 (2012). arXiv:1002.0543 [hepph]
T. Pierzchala, E. RichterWas, Z. Was, M. Worek, Acta Phys. Pol. B 32, 1277 (2001). hepph/0101311
Z. Was, M. Worek, Acta Phys. Pol. B 33, 1875 (2002). hepph/0202007
G.R. Bower, T. Pierzchala, Z. Was, M. Worek, Phys. Lett. B 543, 227 (2002). hepph/0204292
K. Desch, Z. Was, M. Worek, Eur. Phys. J. C 29, 491 (2003). hepph/0302046
K. Desch, A. Imhof, Z. Was, M. Worek, Phys. Lett. B 579, 157 (2004). hepph/0307331
N.D. Christensen, C. Duhr, Comput. Phys. Commun. 180, 1614 (2009). arXiv:0806.4194 [hepph]
J. Alwall, M. Herquet, F. Maltoni, O. Mattelaer, T. Stelzer, J. High Energy Phys. 1106, 128 (2011). arXiv:1106.0522 [hepph]
J. Beringer et al. (Particle Data Group Collaboration), Phys. Rev. D 86, 010001 (2012)
J.H. Kuhn, A. Santamaria, Z. Phys. C 48, 445 (1990)
C. Degrande, C. Duhr, B. Fuks, D. Grellscheid, O. Mattelaer, T. Reiter, Comput. Phys. Commun. 183, 1201 (2012). arXiv:1108.2040 [hepph]
P. de Aquino, W. Link, F. Maltoni, O. Mattelaer, T. Stelzer, Comput. Phys. Commun. 183, 2254 (2012). arXiv:1108.2041 [hepph]
K. Hagiwara, H. Murayama, I. Watanabe, Nucl. Phys. B 367, 257 (1991)
H. Murayama, I. Watanabe, K. Hagiwara, evaluations, KEKReport 9111 (1992)
O. Shekhovtsova, T. Przedzinski, P. Roig, Z. Was, Phys. Rev. D 86, 113008 (2012). arXiv:1203.3955 [hepph]
J.R. Ellis, K.A. Olive, Y. Santoso, New J. Phys. 4, 32 (2002). hepph/0202110
T. Jittoh, K. Kohri, M. Koike, J. Sato, T. Shimomura, M. Yamanaka, Phys. Rev. D 78, 055007 (2008). arXiv:0805.3389 [hepph]
M. Citron, J. Ellis, F. Luo, J. Marrouche, K.A. Olive, K.J. de Vries. arXiv:1212.2886 [hepph]
R.L. Arnowitt, B. Dutta, A. Gurrola, T. Kamon, A. Krislock, D. Toback, Phys. Rev. Lett. 100, 231802 (2008). arXiv:0802.2968 [hepph]
S.S. AbdusSalam, B.C. Allanach, H.K. Dreiner, J. Ellis, U. Ellwanger, J. Gunion, S. Heinemeyer, M. Kraemer et al., Eur. Phys. J. C 71, 1835 (2011). arXiv:1109.3859 [hepph]
B.C. Allanach, Comput. Phys. Commun. 143, 305 (2002). hepph/0104145
C.H. Chen, M. Drees, J.F. Gunion, Phys. Rev. D 55, 330 (1997) [Erratum: Phys. Rev. D 60, 039901 (1999)] [hepph/9607421]
A.J. Barr, C.G. Lester, M.A. Parker, B.C. Allanach, P. Richardson, J. High Energy Phys. 0303, 045 (2003). hepph/0208214
S. Profumo, K. Sigurdson, P. Ullio, M. Kamionkowski, Phys. Rev. D 71, 023518 (2005). astroph/0410714
T. Jittoh, J. Sato, T. Shimomura, M. Yamanaka, Phys. Rev. D 73, 055009 (2006). hepph/0512197
S. Ambrosanio, B. Mele, S. Petrarca, G. Polesello, A. Rimoldi, J. High Energy Phys. 0101, 014 (2001). hepph/0010081
K. Ishiwata, T. Ito, T. Moroi, Phys. Lett. B 669, 28 (2008). arXiv:0807.0975 [hepph]
S. Asai, K. Hamaguchi, S. Shirai, Phys. Rev. Lett. 103, 141803 (2009). arXiv:0902.3754 [hepph]
H. Sugiyama, K. Tsumura, H. Yokoya, Phys. Lett. B 717, 229 (2012). arXiv:1207.0179 [hepph]
Acknowledgements
We wish to thank C. Duhr and P. de Aquino for their help with FeynRules and J. Alwall, T. Stelzer and H. Yokoya for their help with MadGraph5 and Tauola. K.M. also thank J. Keaveney for his help with Root. J.N. acknowledges the warm hospitality extended to him by VUB. T.L. is supported by the ARC Centre of Excellence for Particle Physics at the Terascale. The work presented here has been in part supported by the Concerted Research action “Supersymmetric Models and their Signatures at the Large Hadron Collider”, the Strategic Research Program “High Energy Physics” and the Research Council of the Vrije Universiteit Brussel, in part by the Belgian Federal Science Policy Office through the Interuniversity Attraction Pole P7/37 and IAP VI/11, and by the GrantinAid for Scientific Research (No. 20340064) from the Japan Society for the Promotion of Science.
Open Access
This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License (https://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
About this article
Cite this article
Hagiwara, K., Li, T., Mawatari, K. et al. TauDecay: a library to simulate polarized tau decays via FeynRules and MadGraph5. Eur. Phys. J. C 73, 2489 (2013). https://doi.org/10.1140/epjc/s1005201324894
Received:
Revised:
Published:
DOI: https://doi.org/10.1140/epjc/s1005201324894
Keywords
 Higgs Boson
 Light Supersymmetric Particle
 Helicity Amplitude
 Decay Amplitude
 Collinear Limit