This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.


Probing for chiral ZZ^{\prime} gauge boson through scattering measurement experiments

Kento Asai kento@icrr.u-tokyo.ac.jp Institute for Cosmic Ray Research (ICRR), The University of Tokyo, Kashiwa, Chiba 277-8582, Japan    Arindam Das arindamdas@oia.hokudai.ac.jp Institute for the Advancement of Higher Education, Hokkaido University, Sapporo 060-0817, Japan Department of Physics, Hokkaido University, Sapporo 060-0810, Japan    Jinmian Li jmli@scu.edu.cn College of Physics, Sichuan University, Chengdu 610065, China    Takaaki Nomura nomura@scu.edu.cn College of Physics, Sichuan University, Chengdu 610065, China    Osamu Seto seto@particle.sci.hokudai.ac.jp Department of Physics, Hokkaido University, Sapporo 060-0810, Japan
Abstract

Motivated by the observation of tiny neutrino mass can not be explained within the framework of Standard Model (SM), we consider extra gauge extended scenarios in which tiny neutrino masses are generated through seesaw mechanism. These scenarios are equipped with beyond the standard model (BSM) neutral gauge boson called ZZ^{\prime} in the general U(1)XU(1)_{X} symmetry which is a linear combination of U(1)YU(1)_{Y} and U(1)BLU(1)_{B-L}. In this case, left and right handed fermions interact differently with the ZZ^{\prime}. The ZZ^{\prime} gives rise to different processes involving neutrino-nucleon, neutrino-electron, electron-nucleus and electron-muon scattering processes. By comparing with proton, electron beam-dump experiments data, recast data from searches for the long-lived and dark photon at BaBaR, LHCb and CMS experiments, the electron and muon g2g-2 data, and the data of the dilepton and dijet searches at the LEP experiment, we derive bounds on the gauge coupling and the corresponding gauge boson mass for different U(1)XU(1)_{X} charges and evaluate the prospective limits from the future beam-dump scenarios at DUNE, FASER(2) and ILC. We conclude that large parameter regions could be probed by scattering, beam-dump and collider experiments in future.

preprint: EPHOU-23-013

I Introduction

Tiny neutrino masses and flavor mixing are important experimental observations [1] which motivate us to think beyond the Standard Model (SM). Various cosmological data indicate that nonluminous objects called dark matter capture nearly 0.250.25 fraction of the energy budget of the Universe [2, 3] which further indicate that an extension of the SM is certain. The origin of tiny neutrino mass can be explained by the seesaw mechanism where the SM is extended by SM-singlet Majorana Right Handed Neutrinos (RHNs) [4, 5, 6, 7, 8]. In this case, light neutrino mass can be originated by the suppression of heavy mass scale of the RHNs, which introduces a lepton number violation of unit two which is a nice realization of dimension five Weinberg operator [9].

This simple but interesting ultraviolet (UV) theory is constructed if the SM is extended by a general U(1)XU(1)_{X} gauge group which is a linear combination of U(1)YU(1)_{Y} and U(1)BLU(1)_{B-L}. In this set-up, three generations of SM-singlet RHNs are introduced to cancel the gauge and mixed gauge-gravity anomalies. After the general U(1)XU(1)_{X} symmetry is broken by a SM-singlet scalar, the scalar acquires vacuum expectation value (VEV) which lets the RHNs to acquire Majorana masses. Followed by the electroweak symmetry breaking, a Dirac mass term is generated from the Yukawa interaction between the SM lepton and Higgs doublets along with the SM-singlet RHNs. These Majorana and Dirac masses get involved in the seesaw mechanism to generate light neutrino mass and flavor mixing [10]. All these couplings and interactions are protected by general U(1)XU(1)_{X} gauge symmetry.

In general, an extra gauged U(1)U(1) extension of the SM, a neutral and beyond the SM (BSM) gauge boson, commonly known as the ZZ^{\prime}, exists and acquires the mass after the extra U(1)U(1) symmetry breaking. New physics contributions from the ZZ^{\prime} are considered to be very well motivated and being studied at low and high energy experiments [11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43] from various aspects. A remarkable aspect of our general U(1)XU(1)_{X} models compared with the others studied previously is that the SM left and right chiral fermions could be charged differently under the U(1)XU(1)_{X}. We will study such a chiral scenario to estimate the limits on the general U(1)XU(1)_{X} gauge coupling with respect to the ZZ^{\prime} mass for different general U(1)XU(1)_{X} charges and compare with a variety of existing bounds for the ZZ^{\prime} mass to be probed. In this paper, we consider two cases of general U(1)XU(1)_{X} extensions of the SM. Those two are solutions of anomaly free conditions and mathematically equally possible. In the first case, three generations of RHNs are universal under the general U(1)XU(1)_{X} gauge group having same charge. The second is the less-studied another interesting possibility where first two generations of the RHNs have general U(1)XU(1)_{X} charge 4-4, the third generation RHN has the charge +5+5 and two SU(2)LSU(2)_{L} Higgs doublet are differently charged under the general U(1)XU(1)_{X} gauge group. One of SU(2)LSU(2)_{L} Higgs doublet couples with the RHNs and charged leptons, which is making the scenario neutrinophilic. In addition to the doublet scalars we introduce three SM-singlet scalars which are differently charged under the general U(1)XU(1)_{X} gauge group. We call the second case an alternative general U(1)XU(1)_{X} scenario. Due to the anomaly cancellation conditions, left and right handed SM fermions must be differently charged under the general U(1)XU(1)_{X} gauge group manifesting the chiral nature of the model.

We study the chiral nature of the ZZ^{\prime} interactions through the ZZ^{\prime} mediated neutrino-electron, neutrino-nucleon and electron-muon scattering process at different experiments such as FASERν\nu and FASERν2\nu 2 [44, 45, 46, 47, 48, 49, 50, 51], SND@@LHC [52, 53, 54, 55], NA64 [56, 57, 58, 59, 60, 61], JSNS2, COHERENT, and MuonE [62, 63]. These are the ZZ^{\prime} mediated tt-channel processes and those interaction vertices depending on general U(1)XU(1)_{X} charges manifest the chiral nature which has not been studied before in the literature for these experiments.

In order to compare our results with the existing bounds we obtain respective bounds from different scattering and beam dump experiments in chiral scenario. To do that, we estimate bounds from electron/ positron beam dump experiments Orsay [64], NA64 [65], KEK [66], E141 [67], E137 [68] and E774 [69] respectively. Studying the bounds from the neutrino-electron scattering experiments from TEXONO [70, 71, 72], BOREXINO [73, 74, 75, 76, 77, 78] and JSNS2 at J-PARC experiment [79], we compare our results for different U(1)XU(1)_{X} charges. Depending on the choice of the ZZ^{\prime} mass, we compare our bounds for muon neutrino- and muon anti-neutrino-electron scattering from the CHARM-II experiment [80, 81, 82, 83]. We study neutrino-nucleon scattering to estimate bounds for different U(1)XU(1)_{X} charges from the COHERENT experiment [84, 84, 85, 86, 87, 88]. In addition to those, we compare our results with the GEMMA experiment studying neutrino magnetic moment [89, 27]. We compare the bounds form the dark photon searches at the LHCb experiment [25, 14], dark photon searches from the CMS experiment [90], visible and invisible decay of dark photon from BaBar experiment [91, 92], respectively. We compare our results with proton beam dump experiments like Nomad [93], CHARM [94] and ν\nu-cal [95, 96] experiments respectively. We compare our bounds with the dilepton and dijet final states from the LEP-II study [97, 98, 99, 100]. Finally we estimate bounds on general U(1)XU(1)_{X} coupling with respect to MZM_{Z^{\prime}} from muon and electron g2g-2 experiments to show the complementarity scenario.

Our paper is organized as follows. We discuss the models in Sec. II. We calculate constraints on the gauge coupling fro different U(1)XU(1)_{X} charges in Sec. III. We discuss our results in Sec. IV and finally conclude the paper in Sec. V.

II Model

A general U(1)XU(1)_{X} extension of the SM involves three generations of RHNs to cancel the gauge and mixed gauge-gravity anomalies. As a result, we observe that left and right handed charged fermions in the SM are differently charged under U(1)U(1) group. We write down the UV-complete models in the following:

II.1 Case-I

We consider a general U(1)XU(1)_{X} extension of the SM to investigate the chiral scenario introducing three RHNs (NR)(N_{R}) and an SM-singlet scalar (Φ)(\Phi) field. Three generations of RHNs are introduced to cancel gauge and mixed gauge-gravity anomalies. The corresponding field content is given in Tab. 1 where general U(1)XU(1)_{X} charges, being independent of generations, are written as x~f\tilde{x}_{f} before anomaly cancellation and ff stands for the three generations of quarks (qLα,uRα,dRα)(q_{L}^{\alpha},u_{R}^{\alpha},d_{R}^{\alpha}) and leptons (Lα,eRα,NRα)(\ell_{L}^{\alpha},e_{R}^{\alpha},N_{R}^{\alpha}) respectively where α(=1,2,3)\alpha(=1,~{}2,~{}3) is the generation index. The gauge and mixed gauge-gravity anomaly cancellation conditions in terms of the general charges are written below:

U(1)X[SU(3)C]2\displaystyle{\rm U}(1)_{X}\otimes\left[{\rm SU}(3)_{C}\right]^{2} :\displaystyle\ : 2x~qx~ux~d\displaystyle 2\tilde{x}_{q}-\tilde{x}_{u}-\tilde{x}_{d} = 0,\displaystyle\ =\ 0~{},
U(1)X[SU(2)L]2\displaystyle{\rm U}(1)_{X}\otimes\left[{\rm SU}(2)_{L}\right]^{2} :\displaystyle\ : 3x~q+x~\displaystyle 3\tilde{x}_{q}+\tilde{x}_{\ell} = 0,\displaystyle\ =\ 0~{},
U(1)X[U(1)Y]2\displaystyle{\rm U}(1)_{X}\otimes\left[{\rm U}(1)_{Y}\right]^{2} :\displaystyle\ : x~q8x~u2x~d+3x~6x~e\displaystyle\tilde{x}_{q}-8\tilde{x}_{u}-2\tilde{x}_{d}+3\tilde{x}_{\ell}-6\tilde{x}_{e} = 0,\displaystyle\ =\ 0~{},
[U(1)X]2U(1)Y\displaystyle\left[{\rm U}(1)_{X}\right]^{2}\otimes{\rm U}(1)_{Y} :\displaystyle\ : x~q22x~u2+x~d2x~2+x~e2\displaystyle{\tilde{x}_{q}}^{2}-{\tilde{2x}_{u}}^{2}+{\tilde{x}_{d}}^{2}-{\tilde{x}_{\ell}}^{2}+{\tilde{x}_{e}}^{2} = 0,\displaystyle\ =\ 0~{},
[U(1)X]3\displaystyle\left[{\rm U}(1)_{X}\right]^{3} :\displaystyle\ : 6x~q33x~u33x~d3+2x~3x~ν3x~e3\displaystyle{6\tilde{x}_{q}}^{3}-{3\tilde{x}_{u}}^{3}-{3\tilde{x}_{d}}^{3}+{2\tilde{x}_{\ell}}^{3}-{\tilde{x}_{\nu}}^{3}-{\tilde{x}_{e}}^{3} = 0,\displaystyle\ =\ 0~{},
U(1)X[grav.]2\displaystyle{\rm U}(1)_{X}\otimes\left[{\rm grav.}\right]^{2} :\displaystyle\ : 6x~q3x~u3x~d+2x~x~νx~e\displaystyle 6\tilde{x}_{q}-3\tilde{x}_{u}-3\tilde{x}_{d}+2\tilde{x}_{\ell}-\tilde{x}_{\nu}-\tilde{x}_{e} = 0,\displaystyle\ =\ 0~{}, (1)

respectively. The Yukawa interactions between the fermions and the scalars (H,Φ)(H,\Phi) can be written following the 𝒢SM\mathcal{G}_{\rm SM}\otimes U(1)XU(1)_{X} gauge symmetry as

Yukawa=YuαβqLα¯HuRβYdαβqLα¯H~dRβYeαβLα¯H~eRβYναβLα¯HNRβ12YNαΦ(NRα)c¯NRα+H.c.,{\cal L}^{\rm Yukawa}=-Y_{u}^{\alpha\beta}\overline{q_{L}^{\alpha}}Hu_{R}^{\beta}-Y_{d}^{\alpha\beta}\overline{q_{L}^{\alpha}}\tilde{H}d_{R}^{\beta}-Y_{e}^{\alpha\beta}\overline{\ell_{L}^{\alpha}}\tilde{H}e_{R}^{\beta}-Y_{\nu}^{\alpha\beta}\overline{\ell_{L}^{\alpha}}HN_{R}^{\beta}-\frac{1}{2}Y_{N}^{\alpha}\Phi\overline{(N_{R}^{\alpha})^{c}}N_{R}^{\alpha}+{\rm H.c.}~{}, (2)

where HH is the SM Higgs doublet and we transform it into H~\tilde{H} following iτ2Hi\tau^{2}H^{*} where τ2\tau^{2} is the second Pauli matrix. Hence, using Eq. 2 and following charge neutrality, we express the following relations between the general U(1)XU(1)_{X} charges of the particles as

12xH\displaystyle-\frac{1}{2}x_{H} =\displaystyle= x~q+x~u=x~qx~d=x~x~e=x~+x~ν,\displaystyle-\tilde{x}_{q}+\tilde{x}_{u}\ =\ \tilde{x}_{q}-\tilde{x}_{d}\ =\ \tilde{x}_{\ell}-\tilde{x}_{e}=\ -\tilde{x}_{\ell}+\tilde{x}_{\nu},
2xΦ\displaystyle 2x_{\Phi} =\displaystyle= 2x~ν.\displaystyle-2\tilde{x}_{\nu}~{}. (3)

The general U(1)XU(1)_{X} charges of the fermions can be obtained solving Eqs. (1) and (3) which finally can be expressed using the scalar charges xHx_{H} and xΦx_{\Phi} respectively. Simply the anomaly free charge assignment of the general U(1)XU(1)_{X} can be expressed in terms of a linear combination of two anomaly free scenarios, namely, U(1)YU(1)_{Y} of the SM and BLB-L charges. Finally, we find that the left and right handed fremions under the general U(1)XU(1)_{X} scenario have different charges and hence they interact differently with the neutral, BSM gauge boson ZZ^{\prime} in the model. We take xΦ=1x_{\Phi}=1 without the loss of generality which corresponds to the U(1)BLU(1)_{B-L} and U(1)RU(1)_{R} scenarios with xH=0x_{H}=0 [101, 102, 103, 104, 105] and xH=2x_{H}=2 [106, 107, 108, 109, 110], respectively. U(1)BLU(1)_{B-L} is a vector like scenario where left and right handed fermions of same type are equally charged under the U(1)U(1) extension. In case of U(1)RU(1)_{R} scenario, we find that left handed fermions do not interact with the ZZ^{\prime}.

SU(3)C SU(2)L U(1)Y U(1)X 2-2 1-1 0.5-0.5 0 0.50.5 11 22
U(1)RU(1)_{\rm{R}} U(1)BLU(1)_{B-L}
qLαq_{L}^{\alpha} 3 2 16\frac{1}{6} x~q\tilde{x}_{q} = 16xH+13xΦ\frac{1}{6}x_{H}+\frac{1}{3}x_{\Phi} 0 16\frac{1}{6} 14\frac{1}{4} 13\frac{1}{3} 512\frac{5}{12} 12\frac{1}{2} 13\frac{1}{3}
uRαu_{R}^{\alpha} 3 1 23\frac{2}{3} x~u\tilde{x}_{u} = 23xH+13xΦ\frac{2}{3}x_{H}+\frac{1}{3}x_{\Phi} 1-1 13-\frac{1}{3} 0 13\frac{1}{3} 12\frac{1}{2} 11 53\frac{5}{3}
dRαd_{R}^{\alpha} 3 1 13-\frac{1}{3} x~d\tilde{x}_{d} = 13xH+13xΦ-\frac{1}{3}x_{H}+\frac{1}{3}x_{\Phi} 11 23\frac{2}{3} 12\frac{1}{2} 13\frac{1}{3} 16\frac{1}{6} 0 13-\frac{1}{3}
Lα\ell_{L}^{\alpha} 1 2 12-\frac{1}{2} x~\tilde{x}_{\ell} = 12xHxΦ-\frac{1}{2}x_{H}-x_{\Phi} 0 12-\frac{1}{2} 34-\frac{3}{4} 1-1 54\frac{5}{4} 32-\frac{3}{2} 2-2
eRαe_{R}^{\alpha} 1 1 1-1 x~e\tilde{x}_{e} = xHxΦ-x_{H}-x_{\Phi} 11 0 12-\frac{1}{2} 1-1 32-\frac{3}{2} 2-2 3-3
NRαN_{R}^{\alpha} 1 1 0 x~ν\tilde{x}_{\nu} = xΦ-x_{\Phi} 1-1 1-1 1-1 1-1 1-1 1-1 1-1
HH 1 2 12-\frac{1}{2} x~H\tilde{x}_{H} = 12xH-\frac{1}{2}x_{H} 11 12\frac{1}{2} 14\frac{1}{4} 0 14-\frac{1}{4} 12-\frac{1}{2} 1-1
Φ\Phi 1 1 0 x~Φ\tilde{x}_{\Phi} = 2xΦ2x_{\Phi} 22 22 22 22 22 22 22
Table 1: Field content of general U(1)XU(1)_{X} extension of the SM in the minimal form with the charges of the particles before and after anomaly cancellation considering different benchmark values of xHx_{H} setting xΦ=1x_{\Phi}=1. In this charge assignment, xH=0x_{H}=0 and 2-2 are the U(1)BLU(1)_{B-L} and U(1)RU(1)_{\rm{R}} scenarios. Among the chiral scenarios, U(1)BLU(1)_{B-L} is a vector like case.

Using the general form of the U(1)U(1) charges of the charged fermions, we notice that for xH=1x_{H}=-1 the U(1)U(1) charge of the right handed electron (eR)(e_{R}) becomes zero and, as a result, it has no direct interaction with the ZZ^{\prime} whereas other fermions will interact with the ZZ^{\prime}, manifesting the chiral nature of the model. In similar fashion, we find that for xH=0.5x_{H}=-0.5 the general U(1)XU(1)_{X} charge of the right handed up-type quark uRu_{R} is zero implying no direct interaction with the ZZ^{\prime}, whereas other fermions will have nonzero general U(1)XU(1)_{X} charges allowing direct interactions with the ZZ^{\prime}. Similar behavior could be observed when xH=1x_{H}=1 where general U(1)XU(1)_{X} charge of right handed down type quark (dR)(d_{R}) is zero resulting into no direct interaction with the ZZ^{\prime}. Detailed charge assignments for these combinations of xHx_{H} and xΦx_{\Phi} are given in Tab. 1. We consider two more scenarios with xH=0.5x_{H}=0.5 and 22, setting xΦ=1x_{\Phi}=1, where all the charged fermions interact with the ZZ^{\prime} manifesting the chiral behavior of the model.

The scalar sector of this scenario can be explored by introducing the renormalizable potential of this model and that can be given by

V=mH2(HH)+λH(HH)2+mΦ2(ΦΦ)+λΦ(ΦΦ)2+λmix(HH)(ΦΦ),\displaystyle V\ =\ m_{H}^{2}(H^{\dagger}H)+\lambda_{H}(H^{\dagger}H)^{2}+m_{\Phi}^{2}(\Phi^{\dagger}\Phi)+\lambda_{\Phi}(\Phi^{\dagger}\Phi)^{2}+\lambda_{\rm mix}(H^{\dagger}H)(\Phi^{\dagger}\Phi)~{}, (4)

where HH and Φ\Phi can be separately approximated in the analysis of scalar potential, by taking λmix\lambda_{\rm mix} to be very small. After the breaking of general U(1)XU(1)_{X} and electroweak symmetries, the scalar fields develop vacuum expectation values (VEVs) as follows

H=12(v0),andΦ=vΦ2,\displaystyle\braket{H}\ =\ \frac{1}{\sqrt{2}}\begin{pmatrix}v\\ 0\end{pmatrix}~{},\quad{\rm and}\quad\braket{\Phi}\ =\ \frac{v_{\Phi}}{\sqrt{2}}~{}, (5)

where v=246v=246 GeV is marked as the electroweak scale VEV at the potential minimum and vΦv_{\Phi} is a free parameter. After the general U(1)XU(1)_{X} symmetry is broken, the mass of the BSM neutral gauge boson can be evolved setting xΦ=1x_{\Phi}=1 as

MZ=2gXvΦ,M_{Z^{\prime}}=2g_{X}v_{\Phi}~{}, (6)

in the limit of vΦvv_{\Phi}\gg v. Here, gXg_{X} is the general U(1)XU(1)_{X} coupling and the ZZ^{\prime} mass is a free parameter.

II.2 Case-II

We consider another scenario where the SM is extended by a general U(1)XU(1)_{X} gauge group with three generations of RHNs. The field content of the model is given in Tab. 2. The general U(1)(1) charges of the charged fermion fields are same for all generations, α(=1,2,3)\alpha(=1,2,3) where α\alpha is the generation index. The general U(1)XU(1)_{X} charges of the fields are written as x~f\tilde{x}_{f} before anomaly cancellation and ff stands for the three generations of quarks (qLα,uRα,dRα)(q_{L}^{\alpha},u_{R}^{\alpha},d_{R}^{\alpha}) and leptons (Lα,eRα)(\ell_{L}^{\alpha},e_{R}^{\alpha}) respectively. We introduce two SU(2)SU(2) doublet Higgs fields (H1,2)(H_{1,2}) where one is the SM like (H1)(H_{1}) and the other one is the BSM (H2)(H_{2}) Higgs. Due to different general U(1)XU(1)_{X} charge assignments H1H_{1} does not couple with the BSM fermions. We introduce three SM-singlet scalar fields (Φ1,2,3)(\Phi_{1,2,3}) which are differently charged under general U(1)XU(1)_{X} gauge group. Due to the general U(1)XU(1)_{X} gauge symmetry, first two generations of the RHNs have charge 4-4 each and the third generation RHN has +5+5 charge under the general U(1)XU(1)_{X} gauge group [111]. The RHNs with semi non-universal U(1)U(1) charges in this model help to cancel gauge and mixed gauge-gravity anomalies. As the result, we call this model an alternative general U(1)XU(1)_{X} scenario. Following the gauge and mixed gauge-gravity anomaly cancellation conditions, we relate the general U(1)XU(1)_{X} charges of the charged fermions as

SU(3)C SU(2)L U(1)Y U(1)XU(1)_{X} 2-2 1-1 0.5-0.5 0 0.50.5 11 22
U(1)R(1)_{\rm{R}} B-L
qLαq_{L}^{\alpha} 3 2 16\frac{1}{6} x~q\tilde{x}_{q} = 16xH+13\frac{1}{6}x_{H}+\frac{1}{3} 0 16\frac{1}{6} 14\frac{1}{4} 13\frac{1}{3} 512\frac{5}{12} 12\frac{1}{2} 13\frac{1}{3}
uRαu_{R}^{\alpha} 3 1 23\frac{2}{3} x~u\tilde{x}_{u} = 23xH+13\frac{2}{3}x_{H}+\frac{1}{3} 1-1 13-\frac{1}{3} 0 13\frac{1}{3} 12\frac{1}{2} 11 53\frac{5}{3}
dRαd_{R}^{\alpha} 3 1 13-\frac{1}{3} x~d\tilde{x}_{d} = 13xH+13-\frac{1}{3}x_{H}+\frac{1}{3} 11 23\frac{2}{3} 12\frac{1}{2} 13\frac{1}{3} 16\frac{1}{6} 0 13-\frac{1}{3}
Lα\ell_{L}^{\alpha} 1 2 12-\frac{1}{2} x~\tilde{x}_{\ell} = 12xH1-\frac{1}{2}x_{H}-1 0 12-\frac{1}{2} 34-\frac{3}{4} 1-1 54\frac{5}{4} 32-\frac{3}{2} 2-2
eRαe_{R}^{\alpha} 1 1 1-1 x~e\tilde{x}_{e} = xH1-x_{H}-1 11 0 12-\frac{1}{2} 1-1 32-\frac{3}{2} 2-2 3-3
NR1,2N_{R}^{1,2} 1 1 0 x~ν\tilde{x}_{\nu} = 4-4 4-4 4-4 4-4 4-4 4-4 4-4 4-4
NR3N_{R}^{3} 1 1 0 x~ν\tilde{x}_{\nu}^{\prime} = 55 55 55 55 55 55 55 55
H1H_{1} 1 2 12-\frac{1}{2} x~H1\tilde{x}_{H_{1}} = xH2-\frac{x_{H}}{2} 11 12\frac{1}{2} 14\frac{1}{4} 0 14-\frac{1}{4} 12-\frac{1}{2} 1-1
H2H_{2} 1 2 12-\frac{1}{2} x~H2\tilde{x}_{H_{2}} = 12xH+3-\frac{1}{2}x_{H}+3 44 72\frac{7}{2} 132\frac{13}{2} 33 114\frac{11}{4} 52\frac{5}{2} 22
Φ1\Phi_{1} 1 1 0 x~Φ1\tilde{x}_{\Phi_{1}} = +8+8 +8+8 +8+8 +8+8 +8+8 +8+8 +8+8 +8+8
Φ2\Phi_{2} 1 1 0 x~Φ2\tilde{x}_{\Phi_{2}} = 10-10 10-10 10-10 10-10 10-10 10-10 10-10 10-10
Φ3\Phi_{3} 1 1 0 x~Φ3\tilde{x}_{\Phi_{3}} = 3-3 3-3 3-3 3-3 3-3 3-3 3-3 3-3
Table 2: Field content of the general U(1)XU(1)_{X} extension of the SM in the minimal form with the charges of the particles before and after anomaly cancellation considering different benchmark values of xHx_{H}. Here, xH=0x_{H}=0 is an alternative BLB-L case, which is purely vector like structure given as a reference in this article.
U(1)X[SU(3)C]2\displaystyle{\rm U}(1)_{X}\otimes\left[{\rm SU}(3)_{C}\right]^{2} :\displaystyle\ : 2x~qx~ux~d\displaystyle 2\tilde{x}_{q}-\tilde{x}_{u}-\tilde{x}_{d} = 0,\displaystyle\ =\ 0~{},
U(1)X[SU(2)L]2\displaystyle{\rm U}(1)_{X}\otimes\left[{\rm SU}(2)_{L}\right]^{2} :\displaystyle\ : 3x~q+x~\displaystyle 3\tilde{x}_{q}+\tilde{x}_{\ell} = 0,\displaystyle\ =\ 0~{},
U(1)X[U(1)Y]2\displaystyle{\rm U}(1)_{X}\otimes\left[{\rm U}(1)_{Y}\right]^{2} :\displaystyle\ : x~q8x~u2x~d+3x~6xe~\displaystyle\tilde{x}_{q}-8\tilde{x}_{u}-2\tilde{x}_{d}+3\tilde{x_{\ell}}-6\tilde{x_{e}} = 0,\displaystyle\ =\ 0~{},
[U(1)X]2U(1)Y\displaystyle\left[{\rm U}(1)_{X}\right]^{2}\otimes{\rm U}(1)_{Y} :\displaystyle\ : x~q22x~u2+x~d2x~2+x~e2\displaystyle{\tilde{x}_{q}}^{2}-{2\tilde{x}_{u}}^{2}+{\tilde{x}_{d}}^{2}-{\tilde{x}_{\ell}}^{2}+{\tilde{x}_{e}}^{2} = 0,\displaystyle\ =\ 0~{},
[U(1)X]3\displaystyle\left[{\rm U}(1)_{X}\right]^{3} :\displaystyle\ : 3(6x~q33x~u33x~d3+2x~3x~e3)2x~ν3x~ν3\displaystyle 3({6\tilde{x}_{q}}^{3}-{3\tilde{x}_{u}}^{3}-{3\tilde{x}_{d}}^{3}+{2\tilde{x}_{\ell}}^{3}-{\tilde{x}_{e}}^{3})-2\tilde{x}_{\nu}^{3}-\tilde{x}_{\nu}^{\prime^{3}} = 0,\displaystyle\ =\ 0~{},
U(1)X[grav.]2\displaystyle{\rm U}(1)_{X}\otimes\left[{\rm grav.}\right]^{2} :\displaystyle\ : 3(6x~q3x~u3x~d+2x~x~e)2x~νx~ν\displaystyle 3(6\tilde{x}_{q}-3\tilde{x}_{u}-3\tilde{x}_{d}+2\tilde{x}_{\ell}-\tilde{x}_{e})-2\tilde{x}_{\nu}-\tilde{x}_{\nu}^{\prime} = 0,\displaystyle\ =\ 0~{}, (7)

We find that due to general U(1)XU(1)_{X} charges the SM charged fermions interact differently with the ZZ^{\prime}, manifesting the chiral nature of the model. The second Higgs doublet H2H_{2} interacts with the SM lepton doublet (Lα)(\ell^{\alpha}_{L}) and first two generations of the RHNs (NR1,2)(N_{R}^{1,2}) due to the general U(1)XU(1)_{X} symmetry. Hence, the Dirac Yukawa mass term for NR1,2N_{R}^{1,2} can be generated. On the other hand, the corresponding Majorana mass term for NR1,2N_{R}^{1,2} can be generated from the Dirac Yukawa coupling with Φ1\Phi_{1} followed by the general U(1)XU(1)_{X} symmetry breaking. The third generation of the RHN, NR3N_{R}^{3}, has no Dirac Yukawa coupling involving any of the doublet Higgs fields being prohibited by the general U(1)XU(1)_{X} charge assignments. Therefore, it does not participate in the neutrino mass generation mechanism at the tree level, however, it can have Yukawa interaction with the Φ2\Phi_{2} which further generates a Majorana mass term for NR3N_{R}^{3} after the general U(1)XU(1)_{X} symmetry breaking. Finally, we write the Yukawa interaction among the BSM sector as

int\displaystyle-\mathcal{L}_{\rm int} \displaystyle\ \supset\ α=13β=12Y1αβLα¯H2NRβ+12α=12Y2αΦ1(NRα)c¯NRα+12Y3Φ2(NR3)c¯NR3+H.c.,\displaystyle\sum_{\alpha=1}^{3}\sum_{\beta=1}^{2}Y_{1}^{\alpha\beta}\overline{\ell_{L}^{\alpha}}H_{2}N_{R}^{\beta}+\frac{1}{2}\sum_{\alpha=1}^{2}Y_{2}^{\alpha}\Phi_{1}\overline{(N_{R}^{\alpha})^{c}}N_{R}^{\alpha}+\frac{1}{2}Y_{3}\Phi_{2}\overline{(N_{R}^{3})^{c}}N_{R}^{3}+\rm{H.c.}~{}, (8)

taking Y2Y_{2} being diagonal without the loss of generality. As in the previous case, we can solve the gauge and mixed gauge-gravity anomalies to estimate the charges of the SM particles in Tab. 2.

The scalar potential of this scenario can be given by

V\displaystyle V =\displaystyle\ =\ mH12(H1H1)+λH1(H1H1)2+mH22(H2H2)+λH2(H2H2)2\displaystyle m_{H_{1}}^{2}(H_{1}^{\dagger}H_{1})+\lambda_{H_{1}}(H_{1}^{\dagger}H_{1})^{2}+m_{H_{2}}^{2}(H_{2}^{\dagger}H_{2})+\lambda_{H_{2}}(H_{2}^{\dagger}H_{2})^{2} (9)
+mΦ12(Φ1Φ1)+λ1(Φ1Φ1)2+mΦ22(Φ2Φ2)+λ2(Φ2Φ2)2\displaystyle+m_{\Phi_{1}}^{2}(\Phi_{1}^{\dagger}\Phi_{1})+\lambda_{1}(\Phi_{1}^{\dagger}\Phi_{1})^{2}+m_{\Phi_{2}}^{2}(\Phi_{2}^{\dagger}\Phi_{2})+\lambda_{2}(\Phi_{2}^{\dagger}\Phi_{2})^{2}
+mΦ32(Φ3Φ3)+λ3(Φ3Φ3)2+(μΦ3(H1H2)+H.c.)\displaystyle+m_{\Phi_{3}}^{2}(\Phi_{3}^{\dagger}\Phi_{3})+\lambda_{3}(\Phi_{3}^{\dagger}\Phi_{3})^{2}+(\mu\Phi_{3}(H_{1}^{\dagger}H_{2})+{\rm H.c.})
+λ4(H1H1)(H2H2)+λ5(H1H2)(H2H1)+λ6(H1H1)(Φ1Φ1)\displaystyle+\lambda_{4}(H_{1}^{\dagger}H_{1})(H_{2}^{\dagger}H_{2})+\lambda_{5}(H_{1}^{\dagger}H_{2})(H_{2}^{\dagger}H_{1})+\lambda_{6}(H_{1}^{\dagger}H_{1})(\Phi_{1}^{\dagger}\Phi_{1})
+λ7(H1H1)(Φ2Φ2)+λ8(H1H2)(Φ3Φ3)+λ9(H2H2)(Φ1Φ1)\displaystyle+\lambda_{7}(H_{1}^{\dagger}H_{1})(\Phi_{2}^{\dagger}\Phi_{2})+\lambda_{8}(H_{1}^{\dagger}H_{2})(\Phi_{3}^{\dagger}\Phi_{3})+\lambda_{9}(H_{2}^{\dagger}H_{2})(\Phi_{1}^{\dagger}\Phi_{1})
+λ10(H1H1)(Φ2Φ2)+λ11(H1H2)(Φ3Φ3)+λ12(Φ1Φ1)(Φ2Φ2)\displaystyle+\lambda_{10}(H_{1}^{\dagger}H_{1})(\Phi_{2}^{\dagger}\Phi_{2})+\lambda_{11}(H_{1}^{\dagger}H_{2})(\Phi_{3}^{\dagger}\Phi_{3})+\lambda_{12}(\Phi_{1}^{\dagger}\Phi_{1})(\Phi_{2}^{\dagger}\Phi_{2})
+λ13(Φ2Φ2)(Φ3Φ3)+λ14(Φ3Φ3)(Φ1Φ1).\displaystyle+\lambda_{13}(\Phi_{2}^{\dagger}\Phi_{2})(\Phi_{3}^{\dagger}\Phi_{3})+\lambda_{14}(\Phi_{3}^{\dagger}\Phi_{3})(\Phi_{1}^{\dagger}\Phi_{1})~{}.

Choosing suitable parametrization for the scalar fields in this scenario to develop their respective VEVs, we can write

H1=12(vh10),H2=12(vh20),Φ1=v12,Φ2=v22,Φ3=v32,\displaystyle\braket{H_{1}}\ =\ \frac{1}{\sqrt{2}}\left(\begin{array}[]{c}v_{h_{1}}\\ 0\end{array}\right)~{},\;\braket{H_{2}}\ =\ \frac{1}{\sqrt{2}}\left(\begin{array}[]{c}v_{h_{2}}\\ 0\end{array}\right)~{},\braket{\Phi_{1}}\ =\ \frac{v_{1}}{\sqrt{2}}~{},\;\braket{\Phi_{2}}\ =\ \frac{v_{2}}{\sqrt{2}}~{},\;\braket{\Phi_{3}}\ =\ \frac{v_{3}}{\sqrt{2}}~{},~{}~{}~{}~{} (14)

with the condition, vh12+vh22=246GeV\sqrt{v_{h_{1}}^{2}+v_{h_{2}}^{2}}=246~{}{\rm GeV}. In this alternative general U(1)XU(1)_{X} extension of the SM, we consider negligibly small scalar quartic couplings among SM scalar doublet fields H1,2H_{1,2} and SM-singlet scalar fields Φ1,2,3\Phi_{1,2,3}. As a result, this ensures higher order mixing between the RHNs after the general U(1)XU(1)_{X} breaking to be very strongly suppressed. In Eq. (9), we may consider 0<mmix2=μv3/2mΦ320<m_{\rm mix}^{2}=\mu v_{3}/\sqrt{2}\ll m_{\Phi_{3}}^{2} which further leads to vh2mmix2vh1/mΦ32vh1v_{h_{2}}\sim m_{\rm mix}^{2}v_{h_{1}}/m_{\Phi_{3}}^{2}\ll v_{h_{1}} [112].

Due to the presence of the general U(1)XU(1)_{X} gauge symmetry, the doublet scalar sector H1,2H_{1,2} and singlet scalar sector Φ1,2,3\Phi_{1,2,3} interact only through the coupling Φ3(H1H2)+H.c.\Phi_{3}(H_{1}^{\dagger}H_{2})+{\rm H.c.}, however, this coupling has no significant effect to determine the VEVs (v1,2,3)(v_{1,2,3}) of the singlet scalar fields (Φ1,2,3)(\Phi_{1,2,3}), because there is already one collider constraint present in the form of v12+v22+v32vh12+vh22v_{1}^{2}+v_{2}^{2}+v_{3}^{2}\gg v_{h_{1}}^{2}+v_{h_{2}}^{2}; one find that v12+v22+v32\sqrt{v_{1}^{2}+v_{2}^{2}+v_{3}^{2}} should be typically larger than around 11 TeV from various constraints in the light ZZ^{\prime} case [11] and constraint from dilepton resonance in heavy ZZ^{\prime} case [113, 114] since the value is related to the ZZ^{\prime} boson mass and new gauge coupling. Therefore we arrange the parameters of the scalar potential in a way so that the VEVs of Φ1,2,3\Phi_{1,2,3} will be almost same following μ<v1\mu<v_{1} whereas Φ3\Phi_{3} can be considered as a spurion field which generate the mixing between H1,2H_{1,2} in Eq. (9). Once the Φ3\Phi_{3} acquires the VEV we get the mixing mass term between H1,2H_{1,2} as mmix=μv32m_{\rm mix}=\frac{\mu v_{3}}{\sqrt{2}} which resembles the potential of two Higgs doublet model, however, due to the presence of the general U(1)XU(1)_{X} symmetry the SM-singlet fields Φ1,2,3\Phi_{1,2,3} do not mix. As a result there are two existing physical Nambu-Goldstone (NG) bosons originating from the SM-singlet scalars. Due to the tiny quartic couplings and gauge couplings, the SM-singlet scalars become decoupled from the SM thermal bath in the early universe. Additionally, we consider that the singlet scalars are heavier than the neutral BSM gauge boson ZZ^{\prime} preventing its decay into the NG bosons. The breaking of general U(1) gauge symmetry helps ZZ^{\prime} to acquire the mass which copuld be given by

MZ=gX64v12+100v22+9v32+14xH2vh12+(12xH+3)2vh22gX64v12+100v22+9v32.\displaystyle M_{Z^{\prime}}=g_{X}\sqrt{64v_{1}^{2}+100v_{2}^{2}+9v_{3}^{2}+\frac{1}{4}x_{H}^{2}v_{h_{1}}^{2}+\left(-\frac{1}{2}x_{H}+3\right)^{2}v_{h_{2}}^{2}}\simeq g_{X}\sqrt{64v_{1}^{2}+100v_{2}^{2}+9v_{3}^{2}}~{}. (15)

which is a free parameter and the general U(1)XU(1)_{X} gauge coupling gXg_{X} is also a free parameter. Due to the general U(1)XU(1)_{X} gauge structure H2H_{2} only couples with NR1,2N_{R}^{1,2} making this case a neutrinophilic two Higgs Doublet Model (2HDM) framework [112, 115, 116, 117, 118].

II.3 ZZ^{\prime} interactions with the fermions

After the anomaly cancellation conditions are imposed, we notice that ZZ^{\prime} in the above models can interact with the left and right handed SM fermions differently manifesting chiral nature of the models. Fixing xΦ=1x_{\Phi}=1 in Case-I we find that the chiral nature is the same as the Case-II. Therefore we write the interactions between the fermions with the ZZ^{\prime} in the following as

int=gX(f¯γμqfLPLf+f¯γμqfRPRf)Zμ,\displaystyle\mathcal{L}_{\rm int}=-g_{X}(\overline{f}\gamma_{\mu}q_{f_{L}}P_{L}f+\overline{f}\gamma_{\mu}q_{f_{R}}P_{R}f)Z_{\mu}^{\prime}~{}, (16)

where PL(R)=(1±γ5)/2P_{L(R)}=(1\pm\gamma_{5})/2, qfLq_{f_{L}} and qfRq_{f_{R}} are the corresponding general U(1)(1) charges of the left handed (fL)(f_{L}) and right handed (fR)(f_{R}) fermions. Hence we write vector coupling (cV=qfL+qfR2)(c_{V}=\frac{q_{f_{L}}+q_{f_{R}}}{2}) and axial vector coupling (cA=qfLqfR2)(c_{A}=\frac{q_{f_{L}}-q_{f_{R}}}{2}) for the SM fermions following the charge assignments of the Cases-I and II in Tab. 3 fixing xΦ=1x_{\Phi}=1.

Vector coupling (cV)(c_{V}) Axial-vector coupling (cA)(c_{A})
SM fermions xH=2x_{H}=-2 1-1 0.5-0.5 0 0.50.5 11 22 xH=2x_{H}=-2 1-1 0.5-0.5 0 0.50.5 11 22
Charged lepton (α)(\ell^{\alpha}) 34xH1=12-\frac{3}{4}x_{H}-1=\frac{1}{2} 14-\frac{1}{4} 58-\frac{5}{8} 1-1 118-\frac{11}{8} 74-\frac{7}{4} 52-\frac{5}{2} 14xH=12\frac{1}{4}x_{H}=-\frac{1}{2} 14-\frac{1}{4} 18-\frac{1}{8} 0 18\frac{1}{8} 14\frac{1}{4} 12\frac{1}{2}
SM-like neutrino (νLα)(\nu^{\alpha}_{L}) 14xH+12=0\frac{1}{4}x_{H}+\frac{1}{2}=0 14\frac{1}{4} 38\frac{3}{8} 12\frac{1}{2} 58\frac{5}{8} 34\frac{3}{4} 11 14xH+12=0\frac{1}{4}x_{H}+\frac{1}{2}=0 14\frac{1}{4} 38\frac{3}{8} 12\frac{1}{2} 58\frac{5}{8} 34\frac{3}{4} 11
up-type quarks (quα)(q_{u}^{\alpha}) 512xH+13=12\frac{5}{12}x_{H}+\frac{1}{3}=-\frac{1}{2} 112-\frac{1}{12} 18\frac{1}{8} 13\frac{1}{3} 1324\frac{13}{24} 34\frac{3}{4} 76\frac{7}{6} 14xH=12-\frac{1}{4}x_{H}=-\frac{1}{2} 14-\frac{1}{4} 18-\frac{1}{8} 0 18\frac{1}{8} 14\frac{1}{4} 12\frac{1}{2}
down-type quarks (qdα)(q_{d}^{\alpha}) 112xH+13=12-\frac{1}{12}x_{H}+\frac{1}{3}=-\frac{1}{2} 14-\frac{1}{4} 18-\frac{1}{8} 13\frac{1}{3} 724\frac{7}{24} 14\frac{1}{4} 16\frac{1}{6} 14xH=12-\frac{1}{4}x_{H}=-\frac{1}{2} 14-\frac{1}{4} 18-\frac{1}{8} 0 18\frac{1}{8} 14\frac{1}{4} 12\frac{1}{2}
Table 3: Vector and axial vector couplings in general U(1)XU(1)_{X} scenarios for the couplings between SM fermions and ZZ^{\prime}. In B-L case considering xH=0x_{H}=0 and xΦ=1x_{\Phi}=1, the axial vector couplings for the charged fermions vanish.

Hence, we notice that interactions between SM fermions and ZZ^{\prime} are chiral in nature in general U(1)XU(1)_{X} extension of the SM. The partial decay width of ZZ^{\prime} into different fermions can be calculated using Eq. (16) and we write down the expression as

Γ(Zf¯f)\displaystyle\Gamma(Z^{\prime}\to\bar{f}f) =NCMZgX224π[(qfL2+qfR2)(1mf2MZ2)+6qfLqfRmf2MZ2],\displaystyle=N_{C}\frac{M_{Z^{\prime}}g_{X}^{2}}{24\pi}\left[\left(q_{f_{L}}^{2}+q_{f_{R}}^{2}\right)\left(1-\frac{m_{f}^{2}}{M_{Z^{\prime}}^{2}}\right)+6q_{f_{L}}q_{f_{R}}\frac{m_{f}^{2}}{M_{Z^{\prime}}^{2}}\right]~{}, (17)

where mfm_{f} is the mass of different SM fermions and qL,Rq_{L,R} are the functions of xHx_{H}. Here, NCN_{C} is the color factor being 11 for the SM leptons and 33 for the SM quarks. The light neutrinos (νL)(\nu_{L}) are considered to be massless due to their tiny mass and putting qfR=0q_{f_{R}}=0 in Eq. 17, we obtain the partial decay width of ZZ^{\prime} into a pair of one generation light neutrinos as

Γ(Zνν)=MZgX224πqfL2,\displaystyle\Gamma(Z^{\prime}\to\nu\nu)=\frac{M_{Z^{\prime}}g_{X}^{2}}{24\pi}q_{f_{L}}^{2}~{}, (18)

were qfLq_{f_{L}} is a function of xHx_{H}. The ZZ^{\prime} gauge boson can decay into a pair of heavy Majorana neutrinos if ZZ^{\prime} is heavier than twice the mass of the heavy neutrinos. The corresponding partial decay width into single generation of heavy neutrino pair can be written as

Γ(ZNRαNRα)=mZgX224πqNR2(14mN2MZ2)32,\displaystyle\Gamma(Z^{\prime}\to N_{R}^{\alpha}N_{R}^{\alpha})=\frac{m_{Z^{\prime}}g_{X}^{2}}{24\pi}q_{N_{R}}^{2}\left(1-4\frac{m_{N}^{2}}{M_{Z^{\prime}}^{2}}\right)^{\frac{3}{2}}~{}, (19)

with qNRq_{N_{R}} is the general U(1)XU(1)_{X} charge of the heavy neutrinos which could be found from Tabs. 1 and 2, respectively and mNm_{N} is the mass of the heavy neutrinos. If we consider that the RHNs are heavier than the half of the MZM_{Z^{\prime}} then the decay of ZZ^{\prime} into a pair of RHN is kinematically forbidden. We find that the U(1)XU(1)_{X} charges of the fermions in Cases-II are same as those in Case-I with xΦ=1x_{\Phi}=1. As a result we can utilize same bounds for both the cases.

II.4 Neutrino mass

The Yukawa interactions given in Eqs. 2 and 8 lead us to the generation of neutrino mass mechanism. The general U(1)XU(1)_{X} symmetry breaking generates the Majorana mass term for the three (first two) generations of the RHNs in Case-I (II) where BSM scalar Φ(1)\Phi_{(1)} is involved. The Dirac mass term is generated after the electroweak symmetry breaking where SM Higgs doublet H1(2)H_{1(2)} is involved in Case-I(II). The corresponding Dirac and Majorana mass terms are written in Tab. 4.

Models Majorana mass Dirac mass
(mN)(m_{N}) (mD)(m_{D})
Case-I mNRα=YNα2vΦm_{N_{R}^{\alpha}}=\frac{Y_{N}^{\alpha}}{\sqrt{2}}v_{\Phi} mDαβ=Yν2v1m_{D}^{\alpha\beta}=\frac{Y_{\nu}}{\sqrt{2}}v_{1}
Case-II mNR1,2=Y21,22v1m_{N_{R}^{1,2}}=\frac{Y_{2}^{1,2}}{\sqrt{2}}v_{1} mD1,2=Y11,22vh2m_{D}^{1,2}=\frac{Y_{1}^{1,2}}{\sqrt{2}}v_{h_{2}}
Table 4: Dirac and Majorana masses in the neutrino sector. In Case-II we use the collider constraints to set (v12+v22+v32)(vh12+vh22)(v_{1}^{2}+v_{2}^{2}+v_{3}^{2})\gg(v_{h_{1}}^{2}+v_{h_{2}}^{2}) and in this case first two generations are participating in the neutrino mass generation mechanism.

Finally the light neutrino mass is generated by the seesaw mechanism to explain the origin of tiny neutrino mass term and flavor mixing. Following Tab. 4 the generic formula for the neutrino mass matrix can be written as

mν=(0mDmDTmN).m_{\nu}=\begin{pmatrix}0&m_{D}\\ m_{D}^{T}&m_{N}\end{pmatrix}. (20)

Diaginalizing the neutrino mass matrix we find the light neutrino mass eigenvalues to be `mDmN1mDT`-m_{D}m_{N}^{-1}m_{D}^{T}’. In Case-II NR1,2N_{R}^{1,2} will generate the neutrino mass, on the other hand, at the tree level NR3N_{R}^{3} will not participate in the neutrino mass generation and it can be considered as a potential DM candidate in some scenarios. The neutrino mass generation mechanism and dark matter physics are not main motivations of this work, however, we provide a simple outline in this paper for completeness because these general U(1)XU(1)_{X} scenarios can generate neutrino mass at the tree level from the so-called seesaw mechanism which is an important aspect for studying such scenarios.

III Calculation of the constraints on the chiral gauge couplings

The chiral ZZ^{\prime} gauge boson interacts with the SM fermions, and the couplings depend on the U(1)XU(1)_{X} charge of the SM Higgs doublet. Therefore, experiments for scattering measurements of the SM particles can search the chiral ZZ^{\prime} gauge boson by measuring deviations of the scattering cross section from the SM value. In this section, we show the scattering cross sections contributed by the chiral ZZ^{\prime} gauge boson and methods to calculate constraints on the U(1)XU(1)_{X} gauge coupling. In this work, we consider FASERν(2)\nu(2), SND@@LHC, NA64 and MuonE as the experiments for scattering measurement. In addition, we estimate the constraints from ν\nu-electron, ν\nu-nucleon, electron and proton beam dump, long-lived gauge boson searches, respectively.

III.1 Scattering cross section contributed by chiral ZZ^{\prime} at FASERν(2)\nu(2), SNDD@LHC, NA64 and MuonE

In this subsection, we summarize scattering cross sections via chiral ZZ^{\prime} interactions at FASERν(2)\nu(2), SND@LHC, NA64 and MuonE experiments.

III.1.1 Prospects for FASERν(2)\nu(2) and SND@LHC

The existence of a light ZZ^{\prime} affects the neutral-current deep-inelastic scattering at the LHC far-forward detectors. We study the constraints from the FASERν\nu, FASERν\nu2 and SND@LHC experiments in this subsection. The numbers of neutrinos that pass through the FASERν\nu and SND@LHC detectors have been simulated in Refs. [47, 119]. It has been found that the muon neutrino from pions and kaons decay is most abundant. The energy distributions of which are shown in Fig. 1. The corresponding neutrino flux at the FASERν\nu2 detector can be obtained by rescaling the flux at the FASERν\nu, assuming the neutrino distributes uniformly on the detector surface.

Refer to caption
Refer to caption
Figure 1: Number of muon neutrino pass through the FASERν\nu (left panel) and SND@LHC (right panel) detectors. Thirty energy bins are defined uniformly on the logarithmic scale in [10,10410^{4}] GeV.

The MG5_aMC@NLO package is used to calculate the fixed target deep-inelastic neutrino-proton scattering cross-section at the leading order. Assuming a benchmark detector made of tungsten target, the nCTEQ15FullNuc_184_74 set [120] as implemented in LHAPDF6 [121] is employed as the proton parton distribution function. The partonic collision energy is taken as the factorization and renormalization scales in our simulation. The neutrino-proton scattering cross section σνp\sigma_{\nu p} is related to gXg_{X}, xHx_{H}, ZZ^{\prime} mass as well as the incoming neutrino energy EνE_{\nu}. In Fig. 2, we present the relation between EνE_{\nu} and σνp/Eν\sigma_{\nu p}/E_{\nu} for a few sets of xHx_{H} and MZM_{Z^{\prime}}. The gXg_{X} is fixed to the unity, since the change of gXg_{X} can only lead to a total rescaling. The ZZ^{\prime} contribution is slightly increased with increasing xHx_{H} from 1-1 to 1. The dependence on the MZM_{Z^{\prime}} is more complicated due to the interference effects between the ZZ and ZZ^{\prime} bosons. The ZZ^{\prime} contribution is negligible for MZ100M_{Z^{\prime}}\gtrsim 100 GeV, and the cross-section becomes identical to the SM one in this region.

Refer to caption
Refer to caption
Refer to caption
Figure 2: The neutrino-proton scattering cross section σνp\sigma_{\nu p} divided by the incoming neutrino energy EνE_{\nu}. The gauge coupling gXg_{X} is set to the unity.

Given the neutrino-nucleus cross section (σνN\sigma_{\nu N}, which is 184×σνp184\times\sigma_{\nu p}, mass number of Tungsten is approximately 184), one can estimate the probability of a neutrino interacting with the detector as

P=σνN×Number of NucleiDetector Area=σνNAmdetmN,\displaystyle P=\frac{\sigma_{\nu N}\times~{}\text{Number of Nuclei}}{\text{Detector Area}}=\frac{\sigma_{\nu N}}{A}\frac{m_{\rm det}}{m_{N}}~{},~{} (21)

where NN is the target nucleus, AA is the detector area, and mNm_{N} is the mass of the target nucleus. The relevant detector configurations are listed in Tab. 5.

Detector AA mdetm_{\rm det} Integrated Luminosity
FASERν\nu 25 cm ×\times 25 cm 1.2 tons 150 fb-1
FASERν\nu2 50 cm ×\times 50 cm 10 tons 3000 fb-1
SND@LHC 39 cm ×\times 39 cm 0.8 tons 150 fb-1
Table 5: Detector configurations.

For given gXg_{X}, xHx_{H} and MZM_{Z^{\prime}}, the number of interacting neutrinos in each EνE_{\nu} bin can be calculated by the products of the number of neutrinos passing through the detector and the interaction probability. For illustration, energy spectra for the interacting muon neutrino at the FASERν\nu, FASERν\nu2 and SND@LHC detectors are shown in Fig. 3. The gXg_{X} is taken to be the unity. Three different values of xHx_{H} and six different values of MZM_{Z^{\prime}} are used. The number of interacting neutrinos is increased with increasing xHx_{H} and decreasing MZM_{Z^{\prime}}. Moreover, we can observe that the number at the FASERν\nu detector is around one order of magnitude larger than that at the SND@LHC detector, while it is around two orders of magnitude below that at the FASERν\nu2 detector.

Refer to caption
Refer to caption
Refer to caption
Refer to caption
Refer to caption
Refer to caption
Refer to caption
Refer to caption
Refer to caption
Figure 3: The muon neutrino spectra at the FASERν\nu (top panels), FASERν\nu2 (middle panels) and SND@LHC (bottom panels) detectors.

We follow a similar strategy as proposed in Ref. [33] to estimate the sensitivity reach in the parameter space of gXg_{X} and MZM_{Z^{\prime}} for our model. Only the total numbers of interacting neutrinos are used in defining the χ2\chi^{2} measure (i.e. the shape of EνE_{\nu} spectra are not of concern):

χ2=minα[NBSMνe(1+α)NSMνeNBSMνe+NBSMνμ(1+α)NSMνμNBSMνμ+NBSMντ(1+α)NSMντNBSMντ+(ασnorm)2],\displaystyle\chi^{2}=\min_{\alpha}\left[\frac{N^{\nu_{e}}_{\rm BSM}-(1+\alpha)N^{\nu_{e}}_{\rm SM}}{N^{\nu_{e}}_{\rm BSM}}+\frac{N^{\nu_{\mu}}_{\rm BSM}-(1+\alpha)N^{\nu_{\mu}}_{\rm SM}}{N^{\nu_{\mu}}_{\rm BSM}}+\frac{N^{\nu_{\tau}}_{\rm BSM}-(1+\alpha)N^{\nu_{\tau}}_{\rm SM}}{N^{\nu_{\tau}}_{\rm BSM}}+(\frac{\alpha}{\sigma_{\rm norm}})^{2}\right]~{}, (22)

where NBSMN_{\rm BSM} and NSMN_{\rm SM} are the number of interacting neutrinos of each flavor in our model and in the SM model. The systematic uncertainties (σnorm\sigma_{\rm norm}) in each neutrino flavor are assumed to be the same and only one nuisance parameter α\alpha is used. The χ2\chi^{2} value is obtained by minimizing over the α\alpha. The 95% confidence level sensitivity reach corresponds to χ2=3.84\chi^{2}=3.84.

III.1.2 Constraints from the NA64

The fixed-target experiment NA64 at the CERN SPS [57, 58] aims to search for the ZZ^{\prime} that is produced through the bremsstrahlung process in the high-energy electron beam colliding with heavy nuclei

eZeZZ;Zνν\displaystyle e^{-}Z\to e^{-}ZZ^{\prime};~{}~{}Z^{\prime}\to\nu\nu (23)

where the ZZ^{\prime} is decaying invisibly. The partial width of each ZZ^{\prime} decay channel is calculated by the DARKCAST package [12, 14], assuming that only vector interactions exist. The results are translated into those in our chiral model by using the same method as we proposed in Ref. [10].

The production cross section and the energy spectrum of the ZZ^{\prime} is simulated by the MG5_aMC@NLO package. Since the target nucleus is lead at the NA64 experiment, the nCTEQ15_208_82 set in LHAPDF6 is employed as the proton PDF. The total cross section is proportional to gX2g_{X}^{2}. Taking gX=0.2g_{X}=0.2 and incoming electron beam energy E0=100E_{0}=100 GeV, we present the electron-proton scattering cross sections with respect to varying xHx_{H} and MZM_{Z^{\prime}} in the left panel of Fig. 4. It can be observed that the cross-section is increased with increasing xHx_{H} and decreasing MZM_{Z^{\prime}}. In the right panel of Fig. 4, the normalized bremsstrahlung ZZ^{\prime} spectra are shown. The shape of ZZ^{\prime} spectrum is highly dependent on the ZZ^{\prime} mass while it is almost irrelevant to the xHx_{H} value. Some analytic discussions about the feature of the spectrum are conducted in Ref. [122].

Refer to caption
Refer to caption
Figure 4: Left panel: the total cross section for the bremsstrahlung ZZ^{\prime} production. Right panel: normalized EZ/E0E_{Z^{\prime}}/E_{0} distribution for different ZZ^{\prime} masses. E0=100E_{0}=100 GeV is the electron beam energy.

Having calculated the differential bremsstrahlung ZZ^{\prime} production cross section (dσZ/dEZd\sigma_{Z^{\prime}}/dE_{Z^{\prime}}) as well as the ZZ^{\prime} decay branching ratios for all of the channels, we are able to calculate the number of signal events at the NA64 experiment as follows [56]

nZ(gX,xH,MZ)=0.5EeEeCdσZdEZ[Br(Zνν)+lBr(Zl+l)exp(LECAL+LHCALLZ)],\displaystyle n_{Z^{\prime}}(g_{X},x_{H},M_{Z^{\prime}})=\int_{0.5E_{e}}^{E_{e}}C\frac{d\sigma_{Z^{\prime}}}{dE_{Z^{\prime}}}\left[\text{Br}(Z^{\prime}\to\nu\nu)+\sum_{l}\text{Br}(Z^{\prime}\to l^{+}l^{-})\exp(-\frac{L_{\rm ECAL}+L_{\rm HCAL}}{L_{Z^{\prime}}})\right], (24)

where CC is related to the detector parameters. The LZ=cτZEZ/MZL_{Z^{\prime}}=c\tau_{Z^{\prime}}E_{Z^{\prime}}/M_{Z^{\prime}}, LECALL_{\rm ECAL} and LHCALL_{\rm HCAL} are the ZZ^{\prime} decay length, electromagnetic calorimeter length, and hadronic calorimeter length, respectively. The Ref. [56] uses data with 3.22×10113.22\times 10^{11} electrons on target collected during 2016–2021 runs at the NA64 experiment and obtains the 90% C.L. exclusion limits for the U(1)BLU(1)_{B-L} model. The corresponding bounds for our chiral model with different xHx_{H} can be obtained by requiring the number of signal events nZ(gX,xH,MZ)n_{Z^{\prime}}(g_{X},x_{H},M_{Z^{\prime}}) in Eq. 24 to be the same as that in the B-L case.

III.1.3 Constraints from the MUonE

MUonE [123, 124] is an experiment at CERN aiming to measure elastic scatterings between the 150 GeV μ+\mu^{+} beam and target being electrons in beryllium atoms and determining a contribution of hadronic vacuum polarization to the muon anomalous magnetic moment with a method [125]. The existence of the chiral ZZ^{\prime} changes the cross section of the elastic scattering between positive muon and electron, and therefore, the chiral ZZ^{\prime} can be searched by estimating a deviation of the scattering cross section from the SM prediction.

The elastic scattering cross section between the positive muon and electron is given by

dσ(μ+e)dT=dσ(μ+e)dT|SM+dσ(μ+e)dT|Z+dσ(μ+e)dT|Int,\frac{d\sigma(\mu^{+}e^{-})}{dT}=\left.\frac{d\sigma(\mu^{+}e^{-})}{dT}\right|_{\rm SM}\left.+\frac{d\sigma(\mu^{+}e^{-})}{dT}\right|_{Z^{\prime}}\left.+\frac{d\sigma(\mu^{+}e^{-})}{dT}\right|_{\rm Int}~{}, (25)

where TT denotes the electron recoil energy, and the first, second, and third terms in the right-hand side (RHS) correspond to the contributions from the only SM interaction, only ZZ^{\prime} one and interference between the SM and ZZ^{\prime} ones. The purely SM contribution can be given by

dσ(μ+e)dT|SM=παEM2(Eμ2mμ2)me2(Tme)2{2Eμme(EμT+me)(Tme)(2me2+mμ2meT)},\displaystyle\left.\frac{d\sigma(\mu^{+}e^{-})}{dT}\right|_{\rm SM}=\frac{\pi\alpha_{\rm EM}^{2}}{(E_{\mu}^{2}-m_{\mu}^{2})m_{e}^{2}(T-m_{e})^{2}}\left\{2E_{\mu}m_{e}(E_{\mu}-T+m_{e})-(T-m_{e})(2m_{e}^{2}+m_{\mu}^{2}-m_{e}T)\right\}~{}, (26)

where αEM\alpha_{\rm EM} denotes the fine structure constant, and EμE_{\mu} stands for the energy of the positive muon. The contributions from ZZ^{\prime} exchanging diagram and interference between the SM and ZZ^{\prime} are estimated by

dσ(μ+e)dT|Z=\displaystyle\left.\frac{d\sigma(\mu^{+}e^{-})}{dT}\right|_{Z^{\prime}}= gX4me128πMZ4(Eμ2mμ2)(MZ2+2meTme2)2\displaystyle\frac{g_{X}^{4}m_{e}}{128\pi M_{Z^{\prime}}^{4}(E_{\mu}^{2}-m_{\mu}^{2})(M_{Z^{\prime}}^{2}+2m_{e}T-m_{e}^{2})^{2}}
×[(x~e+x~)4{MZ4(2Eμ(EμT+me)+T23meTmμ2+2me2)}\displaystyle\quad\times\left[(\tilde{x}_{e}+\tilde{x}_{\ell})^{4}\left\{M_{Z^{\prime}}^{4}\left(2E_{\mu}(E_{\mu}-T+m_{e})+T^{2}-3m_{e}T-m_{\mu}^{2}+2m_{e}^{2}\right)\right\}\right.
+2(x~e+x~)3(x~ex~){MZ4(2Eμ2meTmμ2+me2)}\displaystyle\qquad\quad+2(\tilde{x}_{e}+\tilde{x}_{\ell})^{3}(\tilde{x}_{e}-\tilde{x}_{\ell})\left\{M_{Z^{\prime}}^{4}\left(2E_{\mu}^{2}-m_{e}T-m_{\mu}^{2}+m_{e}^{2}\right)\right\}
+2(x~e+x~)2(x~ex~)2{MZ4(2Eμ(Eμ+Tme)(Tme)2)\displaystyle\qquad\quad+2(\tilde{x}_{e}+\tilde{x}_{\ell})^{2}(\tilde{x}_{e}-\tilde{x}_{\ell})^{2}\left\{M_{Z^{\prime}}^{4}\left(2E_{\mu}(E_{\mu}+T-m_{e})-(T-m_{e})^{2}\right)\right.
+2memμ2MZ2(Tme)+2me2mμ2(Tme)2}\displaystyle\hskip 142.26378pt\left.+2m_{e}m_{\mu}^{2}M_{Z^{\prime}}^{2}(T-m_{e})+2m_{e}^{2}m_{\mu}^{2}(T-m_{e})^{2}\right\}
+2(x~e+x~)(x~ex~)3{MZ4(2Eμ2+meT+mμ2me2)+4memμ2MZ2(Tme)\displaystyle\qquad\quad+2(\tilde{x}_{e}+\tilde{x}_{\ell})(\tilde{x}_{e}-\tilde{x}_{\ell})^{3}\left\{M_{Z^{\prime}}^{4}\left(2E_{\mu}^{2}+m_{e}T+m_{\mu}^{2}-m_{e}^{2}\right)+4m_{e}m_{\mu}^{2}M_{Z^{\prime}}^{2}(T-m_{e})\right.
+4me2mμ2(Tme)2}\displaystyle\hskip 142.26378pt\left.+4m_{e}^{2}m_{\mu}^{2}(T-m_{e})^{2}\right\}
+(x~ex~)4{MZ4(2Eμ(EμT+me)+T2meT+mμ2)+4memμ2MZ2(Tme)\displaystyle\qquad\quad+(\tilde{x}_{e}-\tilde{x}_{\ell})^{4}\left\{M_{Z^{\prime}}^{4}\left(2E_{\mu}(E_{\mu}-T+m_{e})+T^{2}-m_{e}T+m_{\mu}^{2}\right)+4m_{e}m_{\mu}^{2}M_{Z^{\prime}}^{2}(T-m_{e})\right.
+4me2mμ2(Tme)2}],\displaystyle\hskip 142.26378pt\left.\left.+4m_{e}^{2}m_{\mu}^{2}(T-m_{e})^{2}\right\}\right]~{}, (27)
dσ(μ+e)dT|Int=\displaystyle\left.\frac{d\sigma(\mu^{+}e^{-})}{dT}\right|_{\rm Int}= αEMgX28(Eμ2mμ2)(Tme)(MZ2+2meTme2)\displaystyle\frac{\alpha_{\rm EM}g_{X}^{2}}{8(E_{\mu}^{2}-m_{\mu}^{2})(T-m_{e})(M_{Z^{\prime}}^{2}+2m_{e}T-m_{e}^{2})}
×[(x~e+x~)2{2Eμ(EμT+me)+T23meTmμ2+2me2}\displaystyle\quad\times\left[(\tilde{x}_{e}+\tilde{x}_{\ell})^{2}\left\{2E_{\mu}\left(E_{\mu}-T+m_{e}\right)+T^{2}-3m_{e}T-m_{\mu}^{2}+2m_{e}^{2}\right\}\right.
+(x~e+x~)(x~ex~)(2Eμ2meTmμ2+me2)+(x~ex~)2(Tme)(2EμT+me)].\displaystyle\qquad\quad\left.+(\tilde{x}_{e}+\tilde{x}_{\ell})(\tilde{x}_{e}-\tilde{x}_{\ell})\left(2E_{\mu}^{2}-m_{e}T-m_{\mu}^{2}+m_{e}^{2}\right)+(\tilde{x}_{e}-\tilde{x}_{\ell})^{2}\left(T-m_{e}\right)\left(2E_{\mu}-T+m_{e}\right)\right].~{}~{} (28)

The number of elastic scattering signals in ii-th bin of the electron recoil energy (Ti<T<Ti+ΔTT_{i}<T<T_{i}+\Delta T) is evaluated by [35]

Ni=TiTi+ΔTdTdσ(μ+e)dTΘ(T)Θ(Tmax(Eμ)T),\displaystyle N_{i}=\mathcal{L}\int_{T_{i}}^{T_{i}+\Delta T}\mathrm{d}T\frac{\mathrm{d}\sigma(\mu^{+}e^{-})}{\mathrm{d}T}\Theta(T)\Theta\left(T_{\rm max}(E_{\mu})-T\right)~{}, (29)

where \mathcal{L} is the integrated luminosity, and =150\mathcal{L}=150 fb-1 for MUonE. In Eq. (29),Θ(T),\Theta(T) is the Heaviside step function, and Tmax(Eμ)T_{\rm max}(E_{\mu}) stands for the maximal value of the electron recoil energy determined by

Tmax(Eμ)=2me(Eμ2mμ2)2Eμme+me2+mμ2.\displaystyle T_{\rm max}(E_{\mu})=\frac{2m_{e}(E_{\mu}^{2}-m_{\mu}^{2})}{2E_{\mu}m_{e}+m_{e}^{2}+m_{\mu}^{2}}~{}. (30)

In Fig. 5, the number of events contributed only by the SM particles and the deviation of the number of events in chiral ZZ^{\prime} model from that in the SM are shown.

Refer to caption
Refer to caption
Refer to caption
Refer to caption
Figure 5: Top-left panel : the distribution of the number of events contributed only by the SM particles in MUonE. Top-right, bottom-left, and bottom-right panels : the distributions of the deviation of the number of events in MUonE from that contributed only by the SM particles : δNi/NiSM(NiSM+ZNiSM)/NiSM\delta N_{i}/N_{i}^{\rm SM}\equiv(N_{i}^{{\rm SM}+Z^{\prime}}-N_{i}^{\rm SM})/N_{i}^{\rm SM} for (MZ/GeV,gX)=(103,103),(101,103),(10,103)(M_{Z^{\prime}}/{\rm GeV},g_{X})=(10^{-3},10^{-3}),(10^{-1},10^{-3}),(10,10^{-3}), respectively.

In this paper, we evaluate the sensitivity of MUonE to the chiral ZZ^{\prime} gauge boson by the χ2\chi^{2}-function, following Ref. [35]. The χ2\chi^{2}-function is calculated by

χ2=i=1100(NiNiSM)2σstat,i2+σsys,i2,\displaystyle\chi^{2}=\sum_{i=1}^{100}\frac{(N_{i}-N_{i}^{\rm SM})^{2}}{\sigma_{{\rm stat},i}^{2}+\sigma_{{\rm sys},i}^{2}}~{}, (31)

where σstat,i=Ni\sigma_{{\rm stat},i}=\sqrt{N_{i}} stands for the statistical uncertainty, and σsys,i=105Ni\sigma_{{\rm sys},i}=10^{-5}N_{i} stands for the systematic uncertainty at the level of 10 ppm [123]. The 95% confidence level sensitivity is obtained by solving χ2=3.84\chi^{2}=3.84.

III.2 Constraints from proton/electron beam dump experiments

For constraints from proton beam dump experiments NOMAD and CHARM, we obtain bound curves by rescaling the bounds of U(1)B-L case given in ref. [11]. The upper bound on {mZ,gX}\{m_{Z^{\prime}},g_{X}\} plane is approximately derived applying scaling [12, 29]

τZ(gBLmax)τZ(gXmax,xH,xΦ),\tau_{Z^{\prime}}(g_{B-L}^{\rm max})\sim\tau_{Z^{\prime}}(g_{X}^{\rm max},x_{H},x_{\Phi})~{}, (32)

where gBLg_{B-L} denotes the gauge coupling in the U(1)B-L case, and τZ\tau_{Z^{\prime}} is the lifetime of the ZZ^{\prime}. The lower bound is also scaled by applying

gXlowgBLlowBR(MZBLγ)BR(ZBLe+e)τ~ZBR(MZγ)BR(Ze+e)τ~ZBL,g_{X}^{\rm low}\sim g_{B-L}^{\rm low}\sqrt{\frac{{\rm BR}(M\to Z^{\prime}_{B-L}\gamma)~{}{\rm BR}(Z^{\prime}_{B-L}\to e^{+}e^{-})\tilde{\tau}_{Z^{\prime}}}{{\rm BR}(M\to Z^{\prime}\gamma)~{}{\rm BR}(Z^{\prime}\to e^{+}e^{-})\tilde{\tau}_{Z^{\prime}_{B-L}}}}~{}, (33)

where τ~\tilde{\tau} is lifetime with gauge coupling being unity, and ZZ^{\prime} is produced via meson decay with M=π0M=\pi^{0} for MONAD and M=ηM=\eta for CHARM dominantly. The ratio of meson decay branching ratio thus represents the ratio of ZZ^{\prime} production cross sections. Here, meson decay branching ratio is estimated using the method given in ref. [12]. In addition to that we use the same rescaling technique to estimate the constraints on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} plane from the FASER [126] and NA62 [127] experiments for different U(1)XU(1)_{X} charges where the ratio of ZZ^{\prime} production cross section between dark photon and U(1)XU(1)_{X} cases is estimated by calculating bremsstrahlung process as a good approximation using method in ref. [10].

For proton beam dump experiment ν\nu-cal, the bremsstrahlung process dominantly produces ZZ^{\prime} boson. In this work we use the excluded region given in ref. [10] on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} space where we take into account chiral structure of the ZZ^{\prime} interactions in estimating the ZZ^{\prime} production cross section.

For electron beam dump constraints from Orsay and KEK, The bound curves are obtained by rescaling the bounds of U(1)B-L case as for the proton beam dump one. We derive the constraint on the upper region on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} plane approximately by scaling with Eq. (32), which is the same as the proton beam dump case. On the other hand, the constraint on the lower region is estimated by [29]

gXlowgBLlow2BR(ZBLe+e)τ~Z(5xH2/4+3xHxΦ+2xΦ2)BR(Ze+e)τ~ZBL,g_{X}^{\rm low}\sim g_{B-L}^{\rm low}\sqrt{\frac{2{\rm BR}(Z^{\prime}_{B-L}\to e^{+}e^{-})\tilde{\tau}_{Z^{\prime}}}{(5x_{H}^{2}/4+3x_{H}x_{\Phi}+2x_{\Phi}^{2}){\rm BR}(Z^{\prime}\to e^{+}e^{-})\tilde{\tau}_{Z^{\prime}_{B-L}}}}~{}, (34)

where ZZ^{\prime} is considered to be produced via bremsstrahlung process.

For electron beam dump constraints from E137 and E141 experiments, we use the results in ref. [10] for excluded region on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} in which chiral structure of the ZZ^{\prime} interactions is taken into account in estimating the ZZ^{\prime} production cross section via bremsstrahlung process.

III.3 Constraints from electron-(anti)neutrino scattering in neutrino experiments

Here we discuss constraints on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} from electron-(anti)neutrino scattering processes that are tested by neutrino experiments: BOREXINO, TEXONO, GEMMA, CHARM-II and the J-PARC Sterile Neutrino Search at the J-PARC Spallation Neutron Source(JSNS22). To obtain the constraints we estimate the electron-(anti)neutrino scattering cross sections under the existence of ZZ^{\prime} interactions. The differential cross section can be expressed as

dσ(νe)dT=dσ(νe)dT|SM+dσ(νe)dT|Z+dσ(νe)dT|Int\frac{d\sigma(\nu e)}{dT}=\left.\frac{d\sigma(\nu e)}{dT}\right|_{\rm SM}+\left.\frac{d\sigma(\nu e)}{dT}\right|_{Z^{\prime}}+\left.\frac{d\sigma(\nu e)}{dT}\right|_{\rm Int} (35)

where TT denotes the electron recoil energy and the first, second and third terms in the RHS correspond to the contributions from the only SM interactions, the only ZZ^{\prime} interactions and interference between the SM and ZZ^{\prime} interactions. The terms in RHS are given as follows [29]. The purely SM contribution can be written by

dσ(νe)dT|SM\displaystyle\left.\frac{d\sigma(\nu e)}{dT}\right|_{\rm SM} =2GF2meπEν2(a12Eν2+a22(EνT)2a1a2meT),\displaystyle=\frac{2G_{F}^{2}m_{e}}{\pi E_{\nu}^{2}}\left(a^{2}_{1}E^{2}_{\nu}+a_{2}^{2}(E_{\nu}-T)^{2}-a_{1}a_{2}m_{e}T\right), (36)

where EνE_{\nu} is the energy of initial neutrino. Here a1a_{1} and a2a_{2} are given by

a1\displaystyle a_{1} ={sin2θW+12,sin2θW,sin2θW12,sin2θW}for{νee,ν¯ee,νβe,ν¯βe},\displaystyle=\left\{\sin^{2}\theta_{W}+\frac{1}{2},\ \sin^{2}\theta_{W},\ \sin^{2}\theta_{W}-\frac{1}{2},\ \sin^{2}\theta_{W}\right\}\ {\rm for}\ \{\nu_{e}e,\bar{\nu}_{e}e,\nu_{\beta}e,\bar{\nu}_{\beta}e\}, (37)
a2\displaystyle a_{2} ={sin2θW,sin2θW+12,sin2θW,sin2θW12}for{νee,ν¯ee,νβe,ν¯βe},\displaystyle=\left\{\sin^{2}\theta_{W},\ \sin^{2}\theta_{W}+\frac{1}{2},\ \sin^{2}\theta_{W},\ \sin^{2}\theta_{W}-\frac{1}{2}\right\}\ {\rm for}\ \{\nu_{e}e,\bar{\nu}_{e}e,\nu_{\beta}e,\bar{\nu}_{\beta}e\}, (38)

where β={μ,τ}\beta=\{\mu,\tau\}. The contribution from ZZ^{\prime} exchanging diagram is estimated by

dσ(ν()αe)dT|Z\displaystyle\left.\frac{d\sigma(\overset{(-)}{\nu}_{\alpha}e)}{dT}\right|_{Z^{\prime}} =gX4(x~)2me4πEν2(2meT+MZ2)[(2Eν22EνT+T2)(b12+b22)±2b1b2(2EνT)TmeT(b12b22)],\displaystyle=\frac{g_{X}^{4}(\tilde{x}_{\ell})^{2}m_{e}}{4\pi E_{\nu}^{2}(2m_{e}T+M^{2}_{Z^{\prime}})}[(2E^{2}_{\nu}-2E_{\nu}T+T^{2})(b^{2}_{1}+b^{2}_{2})\pm 2b_{1}b_{2}(2E_{\nu}-T)T-m_{e}T(b^{2}_{1}-b_{2}^{2})], (39)

where b1=x~+x~e2b_{1}=\frac{\tilde{x}_{\ell}+\tilde{x}_{e}}{2} and b2=x~x~e2b_{2}=\frac{\tilde{x}_{\ell}-\tilde{x}_{e}}{2} with x~,e\tilde{x}_{\ell,e} from Table. 1, and the negative sign of ±\pm is for the process of anti neutrino. The contributions from interference between the SM and ZZ^{\prime} are also written, depending on the process, as follows:

dσ(νee)dT|int=\displaystyle\left.\frac{d\sigma(\nu_{e}e)}{dT}\right|_{\rm int}= GFgX2x~me2πEν2(2meT+MZ2)[2Eν2(b1+b2)+(2Eν22EνT+T2)(b1c1+b2c2)]\displaystyle\frac{G_{F}g_{X}^{2}\tilde{x}_{\ell}m_{e}}{\sqrt{2}\pi E^{2}_{\nu}(2m_{e}T+M^{2}_{Z^{\prime}})}[2E_{\nu}^{2}(b_{1}+b_{2})+(2E^{2}_{\nu}-2E_{\nu}T+T^{2})(b_{1}c_{1}+b_{2}c_{2})]
+T(2EνT)(b1c2+b2c1)meT(b1b2+b1c1b2c2)],\displaystyle+T(2E_{\nu}-T)(b_{1}c_{2}+b_{2}c_{1})-m_{e}T(b_{1}-b_{2}+b_{1}c_{1}-b_{2}c_{2})], (40)
dσ(ν¯ee)dT|int=\displaystyle\left.\frac{d\sigma(\bar{\nu}_{e}e)}{dT}\right|_{\rm int}= GFgX2x~me2πEν2(2meT+MZ2)[2(EνT)2(b1+b2)+(2Eν22EνT+T2)(b1c1+b2c2)]\displaystyle\frac{G_{F}g_{X}^{2}\tilde{x}_{\ell}m_{e}}{\sqrt{2}\pi E^{2}_{\nu}(2m_{e}T+M^{2}_{Z^{\prime}})}[2(E_{\nu}-T)^{2}(b_{1}+b_{2})+(2E^{2}_{\nu}-2E_{\nu}T+T^{2})(b_{1}c_{1}+b_{2}c_{2})]
T(2EνT)(b1c2+b2c1)meT(b1b2+b1c1b2c2)],\displaystyle-T(2E_{\nu}-T)(b_{1}c_{2}+b_{2}c_{1})-m_{e}T(b_{1}-b_{2}+b_{1}c_{1}-b_{2}c_{2})], (41)
dσ(ν()βe)dT|int=\displaystyle\left.\frac{d\sigma(\overset{(-)}{\nu}_{\beta}e)}{dT}\right|_{\rm int}= GFgX2x~me2πEν2(2meT+MZ2)[(2Eν22EνTT2)2(b1c1+b2c1)±T(2EνT)(b1c2+b2c1)]\displaystyle\frac{G_{F}g_{X}^{2}\tilde{x}_{\ell}m_{e}}{\sqrt{2}\pi E^{2}_{\nu}(2m_{e}T+M^{2}_{Z^{\prime}})}[(2E_{\nu}^{2}-2E_{\nu}T-T^{2})2(b_{1}c_{1}+b_{2}c_{1})\pm T(2E_{\nu}-T)(b_{1}c_{2}+b_{2}c_{1})]
meT(b1c1b2c2)],\displaystyle-m_{e}T(b_{1}c_{1}-b_{2}c_{2})], (42)

where c1=1/2+2sin2θWc_{1}=-1/2+2\sin^{2}\theta_{W} and c2=1/2c_{2}=-1/2. We then estimate the differential cross sections and derive the constraints for each experiment in the following way.

BOREXINO: The cross section of νe\nu_{e}-ee scattering process is estimated by the experiment where Eν=862\langle E_{\nu}\rangle=862 keV and T[270,665]T\simeq[270,665] keV for Be7{}^{7}B_{e} solar neutrino. We require the cross section with ZZ^{\prime} interaction should not be more than 8%8\% above that of the SM prediction [74] to obtain the constraint on {MZ,gX}\{M_{Z^{\prime}},g_{X}\}.

TEXONO: ν¯e\bar{\nu}_{e}-ee scattering process is measured by the experiment using 187 kg of CsI(Tl) scintillating crystal array with 29882/7369 kg-day of reactor ON/OFF data with electron recoil energy of T[3,8]T\simeq[3,8] MeV. The χ2\chi^{2} value is estimated as

χ2=bin(RdataRth)2ΔR2,\chi^{2}=\sum_{\rm bin}\frac{(R_{\rm data}-R_{\rm th})^{2}}{\Delta R^{2}}, (43)

where RdataR_{\rm data} and RthR_{\rm th} are the event ratios measured by the experiment and predicted by the cross section in Eq. (35), and ΔR\Delta R is the experimental uncertainty, for each recoil energy bin taken from data in ref. [70]. Here we also applied anti neutrino flux in the reference. The constraint on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} plane is then obtained by χ2\chi^{2} analysis with 90%90\% C.L.

GEMMA: ν¯e\bar{\nu}_{e}-ee scattering is observed with 1.5 kg HPGe detector where energy of neutrino is Eν1\langle E_{\nu}\rangle\sim 1-22 MeV and flux is 2.7×10132.7\times 10^{13} cm-2s-1. The χ2\chi^{2} value is estimated applying the formula Eq. (43) for the data given in ref. [128] with 13000 ON-hours and 3000 OFF-hours, and we derive the upper limit curve on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} with 90%90\% C.L.

CHARM-II: νμ(ν¯μ)\nu_{\mu}(\bar{\nu}_{\mu})-electron scattering is observed where 2677±822677\pm 82 and 2752±882752\pm 88 events are respectively obtained for νμ\nu_{\mu} and ν¯μ\bar{\nu}_{\mu} cases. The mean neutrino energies for νμ\nu_{\mu} and ν¯μ\bar{\nu}_{\mu} are respectively Eνμ=23.7\langle E_{\nu_{\mu}}\rangle=23.7 GeV and Eν¯μ=19.1\langle E_{\bar{\nu}_{\mu}}\rangle=19.1 GeV, and the range of observed recoil energy is 3-24 GeV. The χ2\chi^{2} value is estimated using the formula Eq. (43) for the data given in ref. [82, 83], and we derive the upper curve on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} with 90%90\% C.L.

JSNS22: In the experiment, 3 GeV proton collides with mercury target producing pions giving neutrino beams. We consider νe\nu_{e}-ee and ν¯μ\bar{\nu}_{\mu}-ee scattering processes to obtain the constraint. We estimate number of scattering events applying 3.8×10223.8\times 10^{22} protons on target per year, 17 tons of a gadolinium(Gd)-loaded liquid-scintillator detector (LS) detector and neutrino fluxes given in ref. [79]. The corresponding χ2\chi^{2} is estimated by

χ2=min𝛼[(Nth(1+α)NSM)2Nth+(ασnorm)2],\chi^{2}=\underset{\alpha}{\rm min}\left[\frac{(N_{\rm th}-(1+\alpha)N_{\rm SM})^{2}}{N_{\rm th}}+\left(\frac{\alpha}{\sigma_{\rm norm}}\right)^{2}\right], (44)

where NthN_{\rm th} and NSMN_{\rm SM} are the expected number of events in our models and in the SM for 1 year, σnorm\sigma_{\rm norm} is the systematic uncertainty in the neutrino flux normalization, and α\alpha is nuisance parameter. Here σnorm\sigma_{\rm norm} is assumed to be 5%5\% as the reference value. Then the future sensitivity on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} plane is estimated by requiring χ2\chi^{2} value to be less than that of 90%90\% C.L.

III.4 Limits from coherent neutrino-nucleus scattering

An upper limit curve on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} is also obtained from coherent elastic neutrino-nucleus scattering (CEν\nuNS) that is measured by COHERENT experiment with CsI and Ar targets [129, 87, 130]. Here we derive the curve by rescaling the limit curve for U(1)BLU(1)_{B-L} case given in Refs. [43, 86] by comparing number of events in U(1)BLU(1)_{B-L} and other cases. We estimate the number of events at COHERENT experiment adopting formulas in the reference, as discussed below.

Firstly, the differential cross section for CEν\nuNS process is estimated by [131, 132]

dσνNdT(E,T)=GF2Mπ(1MT2E2)QSM+Z2,\frac{d\sigma_{\nu-N}}{dT}(E,T)=\frac{G_{F}^{2}M}{\pi}\left(1-\frac{MT}{2E^{2}}\right)Q_{{\rm SM}+Z^{\prime}}^{2}, (45)

where EE is the initial neutrino energy, TT is the recoil energy, MM is the mass of target nucleus and QSM+ZQ_{{\rm SM}+Z^{\prime}} is the factor coming from interactions including SM and ZZ^{\prime} gauge bosons. In our models QSM+ZQ_{{\rm SM}+Z^{\prime}} is given by

QSM+Z=(gVp(ν)+2ϵuV+ϵdV)ZFZ(|𝐪2|)+(gVn(ν)+ϵuV+2ϵdV)NFN(|𝐪2|),Q_{{\rm SM}+Z^{\prime}}=\left(g^{p}_{V}(\nu_{\ell})+2\epsilon^{uV}_{\ell\ell}+\epsilon^{dV}_{\ell\ell}\right)ZF_{Z}(|{\bf q}^{2}|)+\left(g^{n}_{V}(\nu_{\ell})+\epsilon^{uV}_{\ell\ell}+2\epsilon^{dV}_{\ell\ell}\right)NF_{N}(|{\bf q}^{2}|), (46)

where gVp(n)g^{p(n)}_{V} is the neutrino-proton(neutron) coupling in the SM, Z(N)Z(N) is the number of proton(neutron) in the target nucleus, and FZ(N)(|𝐪2|)F_{Z(N)}(|{\bf q}^{2}|) is the from factors of the proton(neutron) for the target nucleus. The ϵqV\epsilon^{qV}_{\ell\ell} is effective coupling explicitly given by

ϵqV=gX2x~x~q2GF(𝐪2+MZ2).\epsilon^{qV}_{\ell\ell}=\frac{g_{X}^{2}\tilde{x}_{\ell}\tilde{x}_{q}}{\sqrt{2}G_{F}({\bf q}^{2}+M^{2}_{Z^{\prime}})}. (47)

We adopt the values of gVp(νe)=0.0401g_{V}^{p}(\nu_{e})=0.0401, gVp=0.0318g_{V}^{p}=0.0318 and gVn=0.5094g_{V}^{n}=-0.5094 for the neutrino-proton(neutron) coupling in the SM [133, 134]. The Helm parametrization [135] is applied for the form factors FZ(N)(|𝐪2|)F_{Z(N)}(|{\bf q}^{2}|) using proton rms radii {Rp(Cs),Rp(I),Rp(Ar)}={4.804,4.749,3.448}\{R_{p}({\rm Cs}),R_{p}({\rm I}),R_{p}({\rm Ar})\}=\{4.804,4.749,3.448\} [fm] and neutron rms radii {Rn(Cs),Rn(I),Rn(Ar)}={5.01,4.94,3.55}\{R_{n}({\rm Cs}),R_{n}({\rm I}),R_{n}({\rm Ar})\}=\{5.01,4.94,3.55\} [fm] [136, 137, 138].

Then we adopt the neutrino fluxes for the CEν\nuNS event rate in the experiment which depend on the neutrino fluxes produced from the Spallation Neutron Source (SNS) at the Oak Ridge National Laboratories. They are given by [130, 87]

dNνμdE\displaystyle\frac{dN_{\nu_{\mu}}}{dE} =ηδ(Emπ2mμ22mπ),\displaystyle=\eta\delta\left(E-\frac{m_{\pi}^{2}-m_{\mu}^{2}}{2m_{\pi}}\right), (48)
dNνμ¯dE\displaystyle\frac{dN_{\nu_{\bar{\mu}}}}{dE} =η64E2mμ3(34Emμ),\displaystyle=\eta\frac{64E^{2}}{m_{\mu}^{3}}\left(\frac{3}{4}-\frac{E}{m_{\mu}}\right), (49)
dNνedE\displaystyle\frac{dN_{\nu_{e}}}{dE} =η192E2mμ3(12Emμ),\displaystyle=\eta\frac{192E^{2}}{m^{3}_{\mu}}\left(\frac{1}{2}-\frac{E}{m_{\mu}}\right), (50)

where η=rNPOT/(4πL2)\eta=rN_{\rm POT}/(4\pi L^{2}) with LL, NPOTN_{\rm POT} and rr being respectively the distance between the source and the detector, the number of proton-on-target (POT), and the number of neutrinos per flavor that are produced for each POT. For these values, we use r=9×102r=9\times 10^{-2}, NPOT=13.7×1022N_{\rm POT}=13.7\times 10^{22} and L=27.5L=27.5 m for Ar detector, and r=0.08r=0.08, NPOT=17.6×1022N_{\rm POT}=17.6\times 10^{22} and L=19.5L=19.5 m for CsI detector, respectively.

Finally the theoretical number of events for each energy bin in the experiment is derived from

Ni=N(𝒩)TiTi+1𝑑TA(T)EminEmax𝑑Eν=νe,νμ,νμ¯dNνdEdσνNdT(E,T),N_{i}=N(\mathcal{N})\int^{T_{i+1}}_{T_{i}}dTA(T)\int_{E_{\rm min}}^{E_{\rm max}}dE\sum_{\nu=\nu_{e},\nu_{\mu},\nu_{\bar{\mu}}}\frac{dN_{\nu}}{dE}\frac{d\sigma_{\nu-N}}{dT}(E,T), (51)

where ii corresponds to each recoil energy bin, Emin(max)=MT/2(mμ/2)E_{\rm min(max)}=\sqrt{MT/2}(m_{\mu}/2), and A(T)A(T) is the energy-dependent reconstruction efficiency. We estimate upper limit of the coupling gXg_{X} for each ZZ^{\prime} mass for different models by rescaling that of U(1)BLU(1)_{B-L} case in ref. [86] by comparing the number of events for upper limit of gXg_{X} in U(1)BLU(1)_{B-L} with the number of events in each model.

III.5 Limits from collider experiments: LEP-II, CMS, LHCb and BarBar

Here we briefly summarize our estimation of limit on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} from results of several collider experiments:

(i) The limit from LEP-II: We estimate upper limit curve on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} from the results of LEP-II [139, 100] that measure e+ef¯fe^{+}e^{-}\to\bar{f}f scattering cross sections at the ZZ peak with ff being the SM fermions. The scattering cross sections are estimated including ZZ^{\prime} exchanging diagram in addition to the SM processes. Then the value of cross section is compared with the observed value. As a result, we consider constraints coming from the cross section of e+eqq¯e^{+}e^{-}\to q\bar{q} process giving hadronic final state whose value is σ=41.544±0.037\sigma=41.544\pm 0.037 nb, and e+e+e^{+}e^{-}\to\ell^{+}\ell^{-} process with R=Γhad/Γ=20.768±0.024R_{\ell}=\Gamma_{had}/\Gamma_{\ell}=20.768\pm 0.024. We thus obtain upper limit of gXg_{X} as function of mZm_{Z^{\prime}} when we require the total cross section is within the 90% C.L. of the observed value.

(ii) The limit from dark photon search at the LHC experiments: The bounds on {MZ,gX}\{M_{Z^{\prime}},g_{X}\} are also obtained from the results of CMS [90] and LHCb [140] experiments searching for dark photon AA^{\prime} which decays into μ+μ\mu^{+}\mu^{-} pair. They provide us the bounds on the mass of dark photon mAm_{A^{\prime}} and kinetic mixing parameter ϵ\epsilon. The upper limit on gXg_{X} as a function of MZM_{Z^{\prime}} can be estimated in use of following re-scaling

gXmax(MZ=mA)=ϵmax(mA)eσ(ppA)BR(Aμ+μ)σ(ppZ)BR(Zμ+μ),g_{X}^{\rm max}(M_{Z^{\prime}}=m_{A^{\prime}})=\epsilon^{\rm max}(m_{A^{\prime}})e\sqrt{\frac{\sigma(pp\to A^{\prime})BR(A^{\prime}\to\mu^{+}\mu^{-})}{\sigma(pp\to Z^{\prime})BR(Z^{\prime}\to\mu^{+}\mu^{-})}}, (52)

where σ(ppA(Z))\sigma(pp\to A^{\prime}(Z^{\prime})) is dark photon(ZZ^{\prime}) production cross section estimated by CalcHEP3.5 [141] implementing relevant interactions, and ϵmax(mA)\epsilon^{\rm max}(m_{A^{\prime}}) is the upper limit of kinetic mixing parameter as a function of dark photon mass.

(iii) The limit from BaBar experiment: At the BaBar experiment, dark photon is searched for via the process e+eAγe^{+}e^{-}\to A^{\prime}\gamma [91, 92]. They consider visible AA^{\prime} which decays into {e+e,μ+μ,light mesons}\{e^{+}e^{-},\mu^{+}\mu^{-},\text{light mesons}\} final states and invisible AA^{\prime} which decays into invisible final sate such as neutrinos. To estimate the bounds on gXg_{X}, we rescale the upper limit of gauge coupling in U(1)BLU(1)_{B-L} case given in ref. [11] as function of ZZ^{\prime} mass. The rescaling formula for visible ZZ^{\prime} decay mode is

gXmax(MZ)=gBLmax(MZ)σ(e+eγZBL)BR(ZBLvisible states)σ(e+eγZ)BR(Zvisible states),g_{X}^{\rm max}(M_{Z^{\prime}})=g_{B-L}^{\rm max}(M_{Z^{\prime}})\sqrt{\frac{\sigma(e^{+}e^{-}\to\gamma Z^{\prime}_{B-L})BR(Z^{\prime}_{B-L}\to\text{visible states})}{\sigma(e^{+}e^{-}\to\gamma Z^{\prime})BR(Z^{\prime}\to\text{visible states})}}, (53)

where ZBLZ^{\prime}_{B-L} indicate ZZ^{\prime} boson in case of U(1)BLU(1)_{B-L}. The rescaling formula for invisible ZZ^{\prime} decay mode is

gXmax(MZ)=gBLmax(MZ)σ(e+eγZBL)BR(ZBLν¯ν)σ(e+eγZ)BR(Zν¯ν),g_{X}^{\rm max}(M_{Z^{\prime}})=g_{B-L}^{\rm max}(M_{Z^{\prime}})\sqrt{\frac{\sigma(e^{+}e^{-}\to\gamma Z^{\prime}_{B-L})BR(Z^{\prime}_{B-L}\to\bar{\nu}\nu)}{\sigma(e^{+}e^{-}\to\gamma Z^{\prime})BR(Z^{\prime}\to\bar{\nu}\nu)}}, (54)

where all neutrino modes are taken into account.

III.6 Limits from electron and muon (g2)(g-2)

The ZZ^{\prime} boson contributes to electron and muon (g2)(g-2), Δaμ,e\Delta a_{\mu,e}, at one loop level. Calculating one-loop diagram, we obtain the formula as [142]

Δa=gX28π2m2MZ201𝑑x(x~+x~e)22x2(1x)+(x~x~e)2(2x(1x)(x4)4x3m2MZ2)(1x)(1xm2MZ2)+xm2MZ2,\Delta a_{\ell}=\frac{g_{X}^{2}}{8\pi^{2}}\frac{m_{\ell}^{2}}{M_{Z^{\prime}}^{2}}\int^{1}_{0}dx\frac{(\tilde{x}_{\ell}+\tilde{x}_{e})^{2}2x^{2}(1-x)+(\tilde{x}_{\ell}-\tilde{x}_{e})^{2}\left(2x(1-x)(x-4)-4x^{3}\frac{m_{\ell}^{2}}{M_{Z^{\prime}}^{2}}\right)}{(1-x)\left(1-x\frac{m_{\ell}^{2}}{M_{Z^{\prime}}^{2}}\right)+x\frac{m_{\ell}^{2}}{M_{Z^{\prime}}^{2}}}, (55)

where ={e,μ}\ell=\{e,\mu\}. As a reference we derive the region that accommodate experimental values of electron and muon (g2)(g-2) in the models. For electron (g2)(g-2), the experimentally obtained ranges are

Δae(133Cs)\displaystyle\Delta a_{e}(^{133}\text{Cs}) =(8.8±3.6)×1013[143],\displaystyle=-(8.8\pm 3.6)\times 10^{-13}~{}~{}\text{\cite[cite]{[\@@bibref{Number}{Parker:2018vye}{}{}]}}\,,
Δae(87Rb)\displaystyle\Delta a_{e}(^{87}\text{Rb}) =(4.8±3.0)×1013[144],\displaystyle=(4.8\pm 3.0)\times 10^{-13}~{}~{}\text{\cite[cite]{[\@@bibref{Number}{Morel:2020dww}{}{}]}}\,, (56)

where the corresponding deviation from the SM are 2.4σ2.4\sigma and 1.6σ1.6\sigma respectively. We consider Δae\Delta a_{e} constraint on ZZ^{\prime} interaction to satisfy either range of Δae\Delta a_{e} depending on the charge assignment. For muon (g2)(g-2) we apply the experimentally obtained range of [145]

Δaμ=(25.1±5.9)×1010,\displaystyle\Delta a_{\mu}=(25.1\pm 5.9)\times 10^{-10}, (57)

that corresponds to 4.2σ\sigma deviation from the SM prediction. We show the parameter region that satisfies the Δaμ\Delta a_{\mu} range in the models.

IV Results and Discussions

Refer to caption
Refer to caption
Figure 6: Limits on gXMZg_{X}-M_{Z^{\prime}} plane for xH=2x_{H}=-2 (upper panel, U(1)RU(1)_{R} case) and xH=1x_{H}=-1 (lower panel) taking xΦ=1x_{\Phi}=1 considering 10310^{-3} GeV MZ150\leq M_{Z^{\prime}}\leq 150 GeV. For xH=2x_{H}=-2 we show the region sensitive to the MuonE experiment. Recasting the data we compare the parameter region with those we estimated from LEP, Dark photon searches at BaBaR, LHCb and CMS (CMS Dark) and different beam dump experiments at Orsay, KEK, E137, CHARM, Nomad, ν\nu-cal, E141, E774, NA64, NA62, FASER and involving prospective bounds from FASER(2), DUNE and ILC-BD, respectively from theoretical analyses. For xH=1x_{H}=-1 we show the regions sensitive to FASERν\nu, FASERν\nu2, SND@@LHC, NA64(eN)(eN) and JSNS2 experiments respectively. We compare our results recasting limits from LEP, CHARM-II, GEMMA, BOREXINO, COHERENT, TEXONO, Dark photon search at BaBaR (vis and invis), LHCb and CMS (CMS Dark) and different beam dump experiments at Orsay, KEK, E137, CHARM, Nomad, ν\nu-cal, E141, E774, NA64 and involving prospective bounds from FASER(2), DUNE and ILC, respectively. In this case MUonE bound can not be calculated because eRe_{R} does not interact with the ZZ^{\prime}.

We calculate the exclusion and future sensitivity regions for the chiral ZZ^{\prime} gauge boson from neutrino-nucleon scattering measurements (FASERν\nu(2), SND@LHC), missing energy search (NA64), muon-electron scattering measurements (MUonE), proton and electron beam dump experiments (Orsay, KEK, E137, CHARM, Nomad, ν\nu-cal, E141, E774, NA64), electron-neutrino scattering measurements (BOREXINO, TEXONO, JSNS2), coherent neutrino-nucleus scattering (COHERENT), and collider experiments (LEP-II, CMS, LHCb, BarBar). In Figs. 6-9, we show the exclusion and future sensitivity regions for xH=2,1,0.5,0,0.5,1x_{H}=-2,-1,-0.5,0,0.5,1, and 22 considering xΦ=1x_{\Phi}=1 where the horizontal and vertical axes are ZZ^{\prime} mass within the range 10310^{-3} GeV MZ150\leq M_{Z^{\prime}}\leq 150 GeV and gauge coupling constant gXg_{X}, respectively. The shaded regions have already been excluded, and on the other hand, the curves without shaded regions are sensitivity curves by future experiments. Along this line we compare the sensitivity lines from Supernova (only in case of xH=2x_{H}=-2, bounds for other xHx_{H} can be found in [10]) and beam dump experiments at FASER(2), DUNE and ILC (ILC-BD) from [10], respectively. The shaded regions in the figures are ruled out by respective experiments.

We show the prospective sensitivity in the gXMZg_{X}-M_{Z^{\prime}} plane obtained from the elastic μ+e\mu^{+}-e^{-} scattering at the MUonE experiment in the upper panel of Fig. 6 for xH=2x_{H}=-2 where left handed fermions do not interact with the ZZ^{\prime}. Bounds obtained from the MUonE experiment vary between 6×104gX0.46\times 10^{-4}\leq g_{X}\leq 0.4 for 10310^{-3} GeV MZ150\leq M_{Z^{\prime}}\leq 150 GeV. It is found that MUonE bounds for MZ0.03M_{Z^{\prime}}\leq 0.03 GeV are ruled out by beam dump searches from NA64 and E774 respectively whereas these bounds for MZ0.175M_{Z^{\prime}}\geq 0.175 GeV are ruled out by estimated limits obtained from the dark photon searches at BaBaR, LHCb, CMS (CMS Dark), respectively except for a narrow window from the LHCb experiment around MZ1M_{Z^{\prime}}\simeq 1 GeV. Limits obtained from electron magnetic dipole moment for Cesium-133 is stronger than the limits obtained from μe\mu-e scattering at MUonE for 0.01740.0174 GeV MZ0.0894\leq M_{Z^{\prime}}\leq 0.0894 GeV with 1.4×104gX7.16×1041.4\times 10^{-4}\leq g_{X}\leq 7.16\times 10^{-4}. Following the analysis given in [10] we find that prospective sensitivity from the beam dump experiments at FASER(2) can reach up to 5×1035\times 10^{-3} at MZ=103M_{Z^{\prime}}=10^{-3} GeV whereas ILC-BD reaches up to 2.5×1032.5\times 10^{-3} at that MZM_{Z^{\prime}}. However, these bounds are ruled out by the results from different beam dump searches from NA64, E774 and E141 whereas prospective reach from FASER(2) and ILC-BD can be improved for MZ0.3M_{Z^{\prime}}\simeq 0.3 GeV. Following the analysis in [10] we find that Supernova bounds can be slightly stronger than the ν\nu-cal bounds around MZ0.1M_{Z^{\prime}}\simeq 0.1 GeV. Within 10310^{-3} GeV MZ0.18\leq M_{Z^{\prime}}\leq 0.18 GeV the prospective bounds from DUNE (prospective DUNE sensitivity plots can be found in [10] for 0.10.1 GeV MZ3\leq M_{Z^{\prime}}\leq 3 GeV) is weaker than ν\nu-cal. We find that recent experimental observations from FASER (FASER-exp)[126] and NA62 [127] are represented by gray solid and dot-dashed lines and the corresponding excluded regions are shaded in gray. Most of these limits are well within the ν\nu-cal bounds, however, rest of them are just above the ν\nu-cal contour offering stronger constrains around 0.040.04 GeV MZ0.07\leq M_{Z^{\prime}}\leq 0.07 GeV from FASER-exp and 0.250.25 GeV MZ0.55\leq M_{Z^{\prime}}\leq 0.55 GeV from NA62 experiment respectively. The experimental limits from FASER experiment (FASER-exp) are weaker than the theoretically estimated lines for the FASER experiment represented by blue dotted contour for MZ>0.04M_{Z^{\prime}}>0.04 GeV. Recasting BaBaR(vis) data for dark photon searches we obtain a stringent bound for MZ0.175M_{Z^{\prime}}\leq 0.175 GeV. Recasting the dark photon searches at the LHCb and CMS experiments, we find that stringent bounds can be calculated for 0.1750.175 GeV MZ70\leq M_{Z^{\prime}}\leq 70 GeV and 11 GeV MZ8\leq M_{Z^{\prime}}\leq 8 GeV respectively. Except for small windows in these ranges where MZ10M_{Z^{\prime}}\simeq 10 GeV, stringent bounds can be obtained from the visible final state in BaBaR(vis) experiment. For MZ0.19M_{Z^{\prime}}\leq 0.19 GeV we compare our results recasting the limits from different beam dump experiments including Nomad, CHARM, KEK, E774, E141, E137 and ν\nu-cal experiments with the prospective limits obtained from FASER(2), DUNE and ILC-BD. Furthermore dilepton and dijet searches also rule out the bounds for MZ70M_{Z^{\prime}}\geq 70 GeV. We also show that dilepton and dijet limits when MZM_{Z^{\prime}} is at the ZZ-pole giving a bound about 3×1033\times 10^{-3} when xH=2x_{H}=-2. Due to the structure of the U(1)RU(1)_{R} scenario where left handed fermions do not interact with the ZZ^{\prime}, we find no other constrains in this scenario from experiments like ν\nu-electron, ν\nu-nucleon, etc where we can not explore the coupling between νL\nu_{L} and ZZ^{\prime}.

Limits for xH=1x_{H}=-1 are shown in the lower panel of Fig. 6. For this charge there is no coupling between eRe_{R} and ZZ^{\prime} resulting no direct bound from μ+e\mu^{+}-e^{-} scattering at the MUonE experiment. However, for this value of xHx_{H}, we obtain prospective limits from FASERν(2)\nu(2), SND@@LHC, ν\nu-nucleus scattering at NA64 (NA64(eN)(eN)) where we compare recasting the bounds from the existing scattering experiments like GEMMA, BOREXINO, CHARM-II, TEXONO and COHERENT. Bounds obtained from the TEXNONO and COHERENT experiments are stronger than the other bounds obtained from the existing scattering experiments for the ZZ^{\prime} roughly within 0.030.03 GeV MZ0.1\leq M_{Z^{\prime}}\leq 0.1 GeV. The prospective bounds on the U(1)XU(1)_{X} gauge coupling varies between 7×106gX17\times 10^{-6}\leq g_{X}\leq 1 from the electron-nucleon scattering process in the NA64(eN)(eN) experiment for 10310^{-3} GeV MZ10\leq M_{Z^{\prime}}\leq 10 GeV. However, the bound is comparable with the bounds obtained from the TEXONO and COHERENT experiments for the ZZ^{\prime} mass within 0.030.03 GeV MZ0.1\leq M_{Z^{\prime}}\leq 0.1 GeV. Whereas bounds obtained from NA64(eN)(eN) experiment is weak compared to the beam dump experiments for MZ0.028M_{Z^{\prime}}\leq 0.028 GeV. Here we recast existing results from different beam dump experiments like Orsay, KEK, E137, CHARM, Nomad, ν\nu-cal, E141, E774, NA64, respectively for xH=1x_{H}=-1 to show complementarity with the other experiments. Prospective bounds on the U(1)XU(1)_{X} gauge coupling obtained from the beam dump experiments like FASER(2) and ILC-BD could be stringent compared to the estimated bounds from the existing beam dump experiments for MZ0.2M_{Z^{\prime}}\geq 0.2 GeV. Interestingly the NA64(eN)(eN) line crosses ILC-BD and FASER2 lines at this mass point. Prospective limits obtained by performing the beam dump study at DUNE we obtain that those limits are weaker than the limits obtained by recasting the existing results from the ν\nu-cal experiment. We find that recent experimental observations from FASER (FASER-exp)[126] and NA62 [127] are represented by gray solid and dot-dashed lines and the corresponding excluded regions are shaded in gray. Most of these limits are well within the ν\nu-cal bounds, however, rest of them are just above the ν\nu-cal contour offering stronger constrains around 0.02250.0225 GeV MZ0.08\leq M_{Z^{\prime}}\leq 0.08 GeV from FASER-exp and 0.250.25 GeV MZ0.55\leq M_{Z^{\prime}}\leq 0.55 GeV from NA62 experiment respectively. The FASER-exp contour almost covers the theoretical region shown by the blue dotted line for FASER experiment for MZ>0.07M_{Z^{\prime}}>0.07 GeV. Estimated prospective bounds obtained from ν\nu-nucleon scattering at FASERν\nu, SND@@LHC vary between 6.69×104gX0.296.69\times 10^{-4}\leq g_{X}\leq 0.29 and those in case of FASERν2\nu 2 vary between 2.8×104gX0.142.8\times 10^{-4}\leq g_{X}\leq 0.14 for 0.010.01 GeV MZ150\leq M_{Z^{\prime}}\leq 150 GeV, respectively showing stronger prospective bounds for 91.291.2 GeV MZ150\leq M_{Z^{\prime}}\leq 150 GeV. We recast dilepton and dijet results from LEP which provide most stringent limit as gX5×103g_{X}\simeq 5\times 10^{-3} at the ZZ-pole. We recast recent results of dark photon searches at the LHCb and CMS providing constrains for 0.20.2 GeV MZ70\leq M_{Z^{\prime}}\leq 70 GeV and 11 GeV MZ8\leq M_{Z^{\prime}}\leq 8 GeV respectively. Constrains on gXg_{X} from LHCb vary between 3.12×105gX1.09×1033.12\times 10^{-5}\leq g_{X}\leq 1.09\times 10^{-3} and those from CMS Dark vary between 4.0×104gX1.1×1034.0\times 10^{-4}\leq g_{X}\leq 1.1\times 10^{-3} leaving some narrow windows where BaBaR(vis) provides the strongest bound around MZ10M_{Z^{\prime}}\simeq 10 GeV as gX7×104g_{X}\simeq 7\times 10^{-4}. Studying ν\nu-electron scattering at JSNS2 we find that a prospective sensitivity for gXg_{X} could reach as low as 2.78×1052.78\times 10^{-5} and as high as 3×1033\times 10^{-3} for MZ=0.022M_{Z^{\prime}}=0.022 GeV and 0.220.22 GeV respectively which could be probed in future. JSNS2 provides stronger bound for MZM_{Z^{\prime}} above the ZZ-pole. Recasting the data from the different beam dump experiments we find that prospective bounds obtained from JSNS2 could be weaker than some of these bounds for MZ0.025M_{Z^{\prime}}\leq 0.025 GeV.In addition to that we estimate bounds on the U(1)XU(1)_{X} coupling analyzing the electron g2g-2 data which belongs to the shaded region being ruled out by the scattering, dark photon search and beam dump experiments, respectively.

Refer to caption
Refer to caption
Figure 7: Limits on gXMZg_{X}-M_{Z^{\prime}} plane for xH=0.5x_{H}=-0.5 (upper panel) and xH=0x_{H}=0 (lower panel, B-L case) taking xΦ=1x_{\Phi}=1 considering 10310^{-3} GeV MZ150\leq M_{Z^{\prime}}\leq 150 GeV showing the regions sensitive to FASERν\nu, FASERν\nu2, SND@@LHC, NA64(eN)(eN) and JSNS2 experiments. Recasting the existing results in our case, we compare parameter regions obtained from LEP, dark photon searches at BaBaR(vis and invis), LHCb and CMS(CMS Dark), scattering experiments CHARM-II, GEMMA, BOREXINO, COHERENT, TEXONO and different beam dump experiments from Orsay, KEK, E137, CHARM, Nomad, ν\nu-cal, E141, E774, NA64, NA62, FASER involving prospective bounds from FASER(2), DUNE and ILC(ILC-BD), respectively from theoretical analyses.

Limits for xH=0.5x_{H}=-0.5 are shown in the upper panel of Fig. 7. In this case uRu_{R} does not interact with the ZZ^{\prime}. We obtain the prospective sensitivity on gXg_{X} by estimating μe+\mu^{-}-e^{+} scattering process involving ZZ^{\prime} contribution at MUonE experiment. Estimated limits are weaker than the projected sensitivities obtained from the ν\nu-nucleon scattering at the experiments like FASERν\nu and SND@@LHC except for MZ0.73M_{Z^{\prime}}\leq 0.73 GeV. Studying ν\nu-nucleon scattering for the FASERν2\nu 2 experiment we find that the prospective sensitivities could be stronger than the projections of FASERν\nu, SND@@LHC and MUonE experiments respectively. The prospective limit obtained from the FASERν2\nu 2 experiment could reach at gX0.15g_{X}\simeq 0.15 for MZM_{Z^{\prime}} going beyond the ZZ-pole. At the ZZ-pole we recast dilepton and dijet bounds from LEP experiment and find that the exclusion limit on the coupling could reach at gX3×103g_{X}\simeq 3\times 10^{-3}. Limits estimated by recasting the COHERENT ν\nu-nucleus scattering are obtained to be strong for 0.0660.066 GeV MZ0.213\leq M_{Z^{\prime}}\leq 0.213 GeV where coupling varies between 9.0×105gX2.36×1049.0\times 10^{-5}\leq g_{X}\leq 2.36\times 10^{-4}. We find that COHERENT limits are weaker than the limits obtained by the ee-nucleus scattering at the NA64 experiment (NA64(eN)(eN)) for MZ0.07M_{Z^{\prime}}\leq 0.07 GeV while NA64(eN)(eN) limits are weaker than the those from COHERENT, TEXONO and BOREXINO experiments beyond this MZM_{Z^{\prime}}. Recasting the Dark photon searches at the experiments like BaBRaR, LHCb and CMS we find that stringent limits come from LHCb within 0.210.21 GeV MZ70\leq M_{Z^{\prime}}\leq 70 GeV whereas those from BaBaR(vis) could have severe bounds around MZ10M_{Z^{\prime}}\simeq 10 GeV and CMS Dark could have severe bound around MZ1M_{Z^{\prime}}\simeq 1 GeV and 33 GeV, respectively. We obtain prospective sensitivity from the JSNS2 experiment studying ν\nu-electron scattering where the most stringent bound comes within the range 0.0240.024 GeV MZ0.215\leq M_{Z^{\prime}}\leq 0.215 GeV where the U(1)XU(1)_{X} gauge coupling varies between 2.46×105gX1.97×1042.46\times 10^{-5}\leq g_{X}\leq 1.97\times 10^{-4}. JSNS2 could also produces a stringent prospective bound on gXg_{X} for MZM_{Z^{\prime}} beyond the ZZ-pole which is close to the prospective limit from FASERν2\nu 2 for MZ150M_{Z^{\prime}}\simeq 150 GeV. The prospective sensitivity line from JSNS2 crosses over the bounds obtained from the prospective limits from the beam dump scenarios like FASER at MZ=0.0287M_{Z^{\prime}}=0.0287 GeV whereas FASER2 and ILC-BD lines at MZ=0.0326M_{Z^{\prime}}=0.0326 GeV and the corresponding couplings are 2.8×1052.8\times 10^{-5} and 3.21×1053.21\times 10^{-5}, respectively. We find that recent experimental observations from FASER (FASER-exp)[126] and NA62 [127] are represented by gray solid and dot-dashed lines and the corresponding excluded regions are shaded in gray. Some parts of these limits are well within the ν\nu-cal bounds, however, rest of them are above the ν\nu-cal contour offering stronger constrains around 0.030.03 GeV MZ0.09\leq M_{Z^{\prime}}\leq 0.09 GeV from FASER-exp and 0.2250.225 GeV MZ0.5\leq M_{Z^{\prime}}\leq 0.5 GeV from NA62 experiment respectively. The FASER-exp contour almost covers the theoretical region shown by the blue dotted line for xH=0.5x_{H}=-0.5. Recasting the data obtained from the existing results of E774, E137, E141, NA64, Orsay, KEK, Nomad and ν\nu-cal we find that prospective bounds obtained from DUNE for MZ0.15M_{Z^{\prime}}\leq 0.15 GeV are weaker than the bounds obtained from recasting the data of the ν\nu-cal experiment. We estimated bounds on gXg_{X} from the electron and muon g2g-2 data. However, the limits are weaker compared to different existing and prospective scattering and dark photon experiments mentioned for this charge. Similar behavior is seen for the sensitivity line obtained after recasting the data from GEMMA experiment which is weak compared to existing beam dump, scattering and dark photon searches, respectively.

We estimate limits on gXg_{X} for xH=0x_{H}=0 depending on MZM_{Z^{\prime}} in the lower panel of Fig. 7. We mention that xH=0x_{H}=0 is the well known B-L scenario. We estimate prospective limits at experiments like MUonE, FASERν(2)\nu(2) and SND@@LHC and find that these sensitivities belong to the shaded region below the ZZ-pole. However, above ZZ-pole the prospective limits from FASERν2\nu 2 becomes stringent for MZ150M_{Z^{\prime}}\leq 150 GeV, the range of ZZ^{\prime} mass under consideration. Recasting the dilepton and dijet search results from LEP we find the strongest bound at ZpoleZ-pole to be 0.00180.0018. Recasting available data, we find that limits obtained from GEMMA, CHARM-II, COHERENT also belong to the shaded region being weak compared to the bounds from different scattering experiments involving TEXONO and BOREXINO, beam dump experiments involving E774, E131, E141, NA64, Orsay, KEK, Nomad and ν\nu-cal, CHARM and dark photon search experiments involving BaBaR(visible and invisible modes), LHCb and CMS after we recast the available data from these experiments. We calculate prospective limits from ee-nucleon scattering at the NA64 experiment which are slightly stronger than existing bounds obtained from the TEXONO experiment for 0.020.02 GeV MZ0.164\leq M_{Z^{\prime}}\leq 0.164 GeV where limits on the U(1)XU(1)_{X} coupling can reach up to 2.0×105gX7.2×1042.0\times 10^{-5}\leq g_{X}\leq 7.2\times 10^{-4}. We show bounds obtained from the dark photon search experiments at BaBaR(visible and invisible), LHCb and CMS and found that LHCb provides a strong bound on the U(1)XU(1)_{X} gauge coupling for 0.210.21 GeV MZ70\leq M_{Z^{\prime}}\leq 70 GeV whereas for narrow windows around MZ10M_{Z^{\prime}}\simeq 10 GeV where limits obtained from BaBaR are stringent and limits on gXg_{X} are stringent around MZ1M_{Z^{\prime}}\simeq 1 GeV and 3 GeV from the CMS Dark, respectively. Prospective bounds from the JSNS2 experiment could be stronger than the other scattering experiments for 0.0280.028 GeV MZ0.21\leq M_{Z^{\prime}}\leq 0.21 GeV where constrains on the U(1)XU(1)_{X} coupling constant could be as strong as 1.7×105gX1.3×1041.7\times 10^{-5}\leq g_{X}\leq 1.3\times 10^{-4}. Starting from beyond the ZZ-pole, prospective search reach from JSNS2 could be around gX=0.02g_{X}=0.02 for MZ150M_{Z^{\prime}}\leq 150 GeV. We compare our results with different prospective beam dump scenarios at DUNE, FASER(2) and ILC-BD. Doing that we find that prospective reach from JSNS2 could intersect the prospective beam dump lines from FASER(2) and ILC-BD at {MZ,gX}={0.025GeV,1.55×105},{0.032GeV,1.92×105},{0.035GeV,2.13×105}\{M_{Z^{\prime}},g_{X}\}=\{0.025~{}\rm GeV,1.55\times 10^{-5}\},\{0.032~{}\rm GeV,1.92\times 10^{-5}\},\{0.035~{}\rm GeV,2.13\times 10^{-5}\}, respectively which could be probed in the near future. Prospective limits obtained from the DUNE are weaker than the ν\nu-cal bounds for MZ0.08M_{Z^{\prime}}\leq 0.08 GeV. We find that recent experimental observations from FASER (FASER-exp) [126] and NA62 [127] are represented by gray solid and dot-dashed lines and the corresponding excluded regions are shaded in gray. Some parts of these limits are well within the ν\nu-cal bounds, however, rest of them are above the ν\nu-cal contour offering stronger constrains around 0.030.03 GeV MZ0.08\leq M_{Z^{\prime}}\leq 0.08 GeV from FASER-exp and 0.2250.225 GeV MZ0.5\leq M_{Z^{\prime}}\leq 0.5 GeV from NA62 experiment respectively. The FASER-exp contour covers the theoretical region shown by the blue dotted line for 0.010.01 GeV MZ0.085\leq M_{Z^{\prime}}\leq 0.085 GeV, however, these are stronger than the theoretical limits.

Refer to caption
Refer to caption
Figure 8: Limits on gXMZg_{X}-M_{Z^{\prime}} plane for xH=0.5x_{H}=0.5 (upper panel) and xH=1x_{H}=1(lower panel) taking xΦ=1x_{\Phi}=1 considering 10310^{-3} GeV MZ150\leq M_{Z^{\prime}}\leq 150 GeV showing the regions sensitive to FASERν\nu, FASERν\nu2, SND@@LHC, NA64(eN)(eN) and JSNS2 experiments. Recasting the existing results in our case, we compare parameter regions obtained from scattering experiments at LEP, CHARM-II, GEMMA, BOREXINO, COHERENT, TEXONO, Dark photon searches at BaBaR(vis and invis), LHCb and CMS(CMS Dark) experiments and different beam dump experiments at Orsay, KEK, E137, CHARM, Nomad, ν\nu-cal, E141, E774, NA64, NA62, FASER involving prospective theoretical bounds from FASER(2), DUNE and ILC(ILC-BD), respectively.

The limits on gXMZg_{X}-M_{Z^{\prime}} plane for xH=0.5x_{H}=0.5 has been shown in the upper panel of Fig. 8 where all the fermions interact with the ZZ^{\prime}, however, left and right handed interactions with the ZZ^{\prime} are different. We estimated prospective limits from the μ+e\mu^{+}-e^{-} scattering at the MUonE experiment and compare it with the bounds recasting the data from other scattering experiment involving GEMMA, TEXONO, BOREXONO, CHARM-II and COHERENT. We find that bounds from TEXONO rules out the other results. We also estimate the bounds from the prospective search reaches at the experiments involving SND@@LHC and FASERν(2)\nu(2). We find that FASERν2\nu 2 could provide a stringent limit beyond the ZZ-pole up to MZ150M_{Z^{\prime}}\leq 150 GeV where constrains on gXg_{X} could reach around 0.1. Using the dilepton and dijet searches from LEP we recast those data in our scenario to obtain a strong limit on MZM_{Z^{\prime}} at the ZZ-pole as gX=0.0012g_{X}=0.0012. We find that electron-nucleon scattering in NA64(eN)(eN) gives a strong bound on the U(1)XU(1)_{X} coupling as 2.22×105gX1.17×1042.22\times 10^{-5}\leq g_{X}\leq 1.17\times 10^{-4} for the ZZ^{\prime} within 0.0280.028 GeV MZ0.14\leq M_{Z^{\prime}}\leq 0.14 GeV which is slightly stronger than the bounds obtained from TEXONO and COHERENT within that range of MZM_{Z^{\prime}}. We calculate bounds on gXg_{X} recasting the data obtained from the dark photon searches at BaBaR, LHCb and CMS, respectively. Hence we find that LHCb provides a stringent bound for the ZZ^{\prime} mass within 0.210.21 GeV MZ70\leq M_{Z^{\prime}}\leq 70 GeV where limits on gXg_{X} vary between 105gX2×10410^{-5}\leq g_{X}\leq 2\times 10^{-4}, while, around MZ10M_{Z^{\prime}}\simeq 10 GeV BaBaR provides stringent constraints whereas around MZ1M_{Z^{\prime}}\simeq 1 GeV and 33 GeV CMS Dark provides stringent constraints on the U(1)XU(1)_{X} gauge coupling. We study νe\nu-e scattering for JSNS2 experiment. JSNS2 shows prospective strongest bounds compared to other scattering experiments and dark photon searches within 0.0310.031 GeV MZ0.21\leq M_{Z^{\prime}}\leq 0.21 GeV where limits on the coupling vary between 1.32×105gX9.1×1051.32\times 10^{-5}\leq g_{X}\leq 9.1\times 10^{-5}. Following the JSNS2 line we find that it may provide a strong bound on gXg_{X} beyond the ZZ-pole which might reach up to gX0.06g_{X}\simeq 0.06 for MZ150M_{Z^{\prime}}\leq 150 GeV. Limits on gXg_{X} calculating electron and muon g2g-2 scenarios are found to be weak compared to the existing scattering, beam dump and dark photon search experiments. Recasting the data from the beam dump experiments involving NA64, E141, E137, CHARM, Nomad, ν\nu-cal, KEK, we find the limits on gXMZg_{X}-M_{Z^{\prime}} plane shown by the shaded areas. We find that recent experimental observations from FASER (FASER-exp) [126] and NA62 [127] are represented by gray solid and dot-dashed lines and the corresponding excluded regions are shaded in gray. Some parts of these limits are well within the ν\nu-cal bounds, however, rest of them are above the ν\nu-cal contour offering stronger constrains around 0.0350.035 GeV MZ0.09\leq M_{Z^{\prime}}\leq 0.09 GeV from FASER-exp and 0.2250.225 GeV MZ0.525\leq M_{Z^{\prime}}\leq 0.525 GeV from NA62 experiment respectively. We also show the prospective sensitivities from FASER(2), DUNE and ILC-BD where sensitivity from DUNE is weaker compared to ν\nu-cal for MZ0.08M_{Z^{\prime}}\leq 0.08 GeV. The prospective sensitivity line from JSNS2 crosses the prospective FASER, FASER2 and ILC-BD lines at {MZ,gX}={0.0305GeV,1.29×105},{0.035GeV,1.54×105},{0.0374GeV,1.65×105}\{M_{Z^{\prime}},g_{X}\}=\{0.0305\rm{GeV},1.29\times 10^{-5}\},~{}\{0.035\rm{GeV},1.54\times 10^{-5}\},~{}\{0.0374\rm{GeV},1.65\times 10^{-5}\}, respectively. These limits could be probed by the scattering and beam dump experiments in future.

In the lower panel of Fig. 8 we show the constrains on the general U(1)XU(1)_{X} coupling for different MZM_{Z^{\prime}} using xH=1x_{H}=1 where dRd_{R} does not interact with the ZZ^{\prime}. We estimate prospective limits from μ+e\mu^{+}-e^{-} scattering at the MUonE experimet and compare it with the bounds recasting the results from CHARM-II, TEXONO, BOREXINO, GEMMA, and COHERENT experiments, respectively. We find that comparing with all these limits MUonE is weak staying in the shaded region for 10310^{-3} GeV MZ150\leq M_{Z^{\prime}}\leq 150 GeV. Prospective sensitivity obtained from the electron-nucleon scattering at the NA64 experiment are denoted by NA64(eN)(eN) line. This provides a strongest prospective limit within 0.02570.0257 GeV MZ0.0085\leq M_{Z^{\prime}}\leq 0.0085 GeV where U(1)XU(1)_{X} coupling could reach down to 1.423×105gX1051.423\times 10^{-5}\leq g_{X}\leq 10^{-5}. Recasting the TEXONO data we find that in this context strong constrains come for 0.02540.0254 GeV MZ0.072\leq M_{Z^{\prime}}\leq 0.072 GeV where limits on the gauge coupling vary within 1.72×105gX4.64×1051.72\times 10^{-5}\leq g_{X}\leq 4.64\times 10^{-5}. We find that bounds obtained from TEXONO can be stronger than the limits obtained from recasting the dilepton and dijet searches at the LEP experiment beyond the ZZ- pole but MZ150M_{Z^{\prime}}\leq 150 GeV where limits on the gauge coupling could vary between 0.031gX0.0510.031\leq g_{X}\leq 0.051. In addition, we obtain that LEP bounds at ZZ-pole could reach at gX0.0019g_{X}\simeq 0.0019 from the dilepton and dijet searches. FASERν2\nu 2 provides a prospective sensitivity beyond the ZZ-pole and below 150 GeV where limits could vary between 0.03gX0.050.03\leq g_{X}\leq 0.05. The fermions in the U(1)XU(1)_{X} scenario under consideration interact equally with the ZZ^{\prime} making a generation independent nature of the model irrespective of xHx_{H} which affects bounds from g2g-2 analysis. This is true for any value of xHx_{H}. We find that limits obtained from the dark photon searches at LHCb shows the strongest bounds on U(1)XU(1)_{X} gauge coupling within 8.41×106gX1.637×1048.41\times 10^{-6}\leq g_{X}\leq 1.637\times 10^{-4} for 0.210.21 GeV MZ70\leq M_{Z^{\prime}}\leq 70 GeV. We find that dark photon search at CMS provides the strongest bounds on the gauge coupling for MZM_{Z^{\prime}} around 1 GeVa nd 3 GeV respectively whereas same scenario appears from the BaBaR experiment for MZM_{Z^{\prime}} around 10 GeV. Prospective limits estimated in the context of JSNS2 experiment are found to reach at 1.85×104gX1.95×1041.85\times 10^{-4}\leq g_{X}\leq 1.95\times 10^{-4} for 0.5170.517 GeV MZ0.56\leq M_{Z^{\prime}}\leq 0.56 GeV and beyond ZZ-pole, the bounds are comparable with the prospective ones from FASERν2\nu 2. We find that recent experimental observations from FASER (FASER-exp) [126] and NA62 [127] are represented by gray solid and dot-dashed lines and the corresponding excluded regions are shaded in gray. Some parts of these limits are well within the ν\nu-cal bounds, however, rest of them are above the ν\nu-cal contour offering stronger constrains around 0.0350.035 GeV MZ0.09\leq M_{Z^{\prime}}\leq 0.09 GeV from FASER-exp and 0.2250.225 GeV MZ0.525\leq M_{Z^{\prime}}\leq 0.525 GeV from NA62 experiment respectively. Prospective limits from JSNS2 crosses the prospective sensitivity lines at the beam dump experiments involving FASER, FASER2 and ILC-BD at {MZ,gX}={0.033GeV,1.05×105},{0.036GeV,1.21×105},{0.0375GeV,1.30×105}\{M_{Z^{\prime}},g_{X}\}=\{0.033\rm{GeV},1.05\times 10^{-5}\},~{}\{0.036\rm{GeV},1.21\times 10^{-5}\},~{}\{0.0375\rm{GeV},1.30\times 10^{-5}\}, respectively which could be probed in future. We compare these bounds recasting the data from different beam dump experiments like E141, NA64, KEK, Orsay, CHARM, E137 and ν\nu-cal which are shown by different shaded regions. We find that prospective bounds obtained from DUNE are weaker than ν\nu-cal for MZ0.06M_{Z^{\prime}}\leq 0.06 GeV.

Refer to caption
Figure 9: Limits on gXMZg_{X}-M_{Z^{\prime}} plane for xH=2x_{H}=2 taking xΦ=1x_{\Phi}=1 considering 10310^{-3} GeV MZ150\leq M_{Z^{\prime}}\leq 150 GeV showing the regions sensitive to FASERν\nu, FASERν\nu2, SND@@LHC, NA64(eN)(eN) and JSNS2 experiments. Recasting the existing results in our case, we compare parameter regions obtained from the scattering experiments at LEP, CHARM-II, GEMMA, BOREXINO, COHERENT, TEXONO, dark photon searches at BaBaR(vis and invis), LHCb and CMS(CMS Dark) and different beam dump experiments at Orsay, KEK, E137, CHARM, Nomad, ν\nu-cal, E141, E774, NA64, NA62 and FASER and involving prospective bounds from FASER(2), DUNE and ILC(ILC-BD) respectively.

We show the bounds from different experiments for xH=2x_{H}=2 in Fig. 9. This charge is another example where all fermions interact with the ZZ^{\prime}, however, their left and right handed counterparts interact differently with the ZZ^{\prime}. Being influenced by the charge assignment we find that prospective bounds on gXg_{X} for different MZM_{Z^{\prime}} obtained form the MUonE experiment are weaker compared to the bounds after recasting the data from TEXONO, BOREXINO, CHARM-II and COHERENT experiments. We estimate prospective bounds on gXg_{X} for different prospective experiments like SND@@LHC and FASERν(2)\nu(2). We find that prospective searches from FASERν2\nu 2 are strong only beyond the ZZ-pole and up to MZ150M_{Z^{\prime}}\leq 150GeV. In this mass range the limits on gXg_{X} vary between 0.014gX0.0250.014\leq g_{X}\leq 0.025. Recasting the dilepton and dijet searches from the LEP experiment we find that limit on gXg_{X} at the ZZ-pole could be as stringent as 6.8×1046.8\times 10^{-4}. We also find that electron-nucleon scattering in NA64 experiment shown by the line NA64(eN)(eN) provides a strong bound on the gauge coupling around 1.1×105gX1.1×1041.1\times 10^{-5}\leq g_{X}\leq 1.1\times 10^{-4} for 0.02610.0261 GeV MZ0.21\leq M_{Z^{\prime}}\leq 0.21 GeV. Studying ν\nu-electron scattering in context of JSNS2 experiment we estimate the prospective bounds on gXg_{X} with respect to MZM_{Z^{\prime}}. The strongest future limits on gXg_{X} can be estimated for ZZ^{\prime} within 0.0320.032 GeV MZ0.212\leq M_{Z^{\prime}}\leq 0.212 GeV as 7.735×106gX5.13×1057.735\times 10^{-6}\leq g_{X}\leq 5.13\times 10^{-5} which crosses respective future sensitivity lines obtained from the beam dump experiments like FASER, FASER2 and ILC-BD at {MZ,gX}={0.033GeV,7.855×106},{0.0371GeV,8.57×106},{0.0386GeV,9.07×106}\{M_{Z^{\prime}},g_{X}\}=\{0.033~{}\rm{GeV},7.855\times 10^{-6}\},~{}\{0.0371~{}\rm{GeV},8.57\times 10^{-6}\},~{}\{0.0386~{}\rm{GeV},9.07\times 10^{-6}\} which could be probed in future. We compare our results for xH=2x_{H}=2 recasting the bounds obtained from the existing results from the beam dump experiments like E141, NA64, KEK, Orsay, CHARM, E137 and ν\nu-cal respectively. The excluded regions are shown by different shades. We find that recent experimental observations from FASER (FASER-exp) [126] and NA62 [127] are represented by gray solid and dot-dashed lines and the corresponding excluded regions are shaded in gray. Some parts of these limits are well within the ν\nu-cal bounds, however, rest of them are above the ν\nu-cal contour offering stronger constrains around 0.0350.035 GeV MZ0.095\leq M_{Z^{\prime}}\leq 0.095 GeV from FASER-exp and 0.2250.225 GeV MZ0.525\leq M_{Z^{\prime}}\leq 0.525 GeV from NA62 experiment respectively. We find that prospective bounds obtained from the beam dump scenario at DUNE for MZ0.04M_{Z^{\prime}}\leq 0.04 GeV are weaker than those obtained from ν\nu-cal experiment recasting the existing data. Dark photon searches from LHCb provides stronger limit for 0.210.21 GeV MZ70\leq M_{Z^{\prime}}\leq 70 GeV and beyond ZZ-pole up to MZ=150M_{Z^{\prime}}=150 GeV. We find that LHCb limits below ZZ-pole vary within 3.5×106gX1043.5\times 10^{-6}\leq g_{X}\leq 10^{-4}. Stringent limits can be obtained from the dark photon search experiments at BaBaR and CMS Dark around MZ10M_{Z^{\prime}}\simeq 10 GeV and MZ1M_{Z^{\prime}}\simeq 1 GeV and 33 GeV within narrow windows from LHCb experiment. Finally we comment that limits obtained from muon and electron g2g-2 studies are weaker than the scattering and beam dump experiments due to the generation independent nature of the fermionic couplings with ZZ^{\prime}.

V Conclusions

In this paper we consider chiral scenarios where ZZ^{\prime} interacts with the left and right handed fermions differently. We obtain that depending on U(1)XU(1)_{X} charges the interactions of the femions with the ZZ^{\prime} get modified by manifesting chiral nature of the scenarios under consideration. Such interactions affect ZZ^{\prime} mediated neutrino-electron, electron-nucleon, electron-muon scattering processes which could be probed at the experiments like FASERν(2)\nu(2), SND@@LHC, NA64(eN)(eN), MUonE, JSNS2, dark photon searches at BaBaR, LHCb and CMS experiments respectively. Further we compare our results with dilepton, dijet searches from LEP, neutrino-nucleus coherent scattering at COHERENT experiment, electron-neutrino scattering experiments like BOREXINO, TEXONO, GEMMA and CHARM-II respectively. We compare our results studying visible and invisible final states at the BaBaR experiment. Finally we show complementarity with different beam dump experiments like ν\nu-cal, E137, E141, NA64, E774, Orsay, CHARM, KEK, Nomad and future experiments like FASER, FASER2 and ILC-BD. We find that the experimental results from FASER matches with the theoretical limits estimated for general U(1)XU(1)_{X} charges having some bounds stronger than our estimated ones for increasing MZM_{Z^{\prime}}. We have also shown the NA62 regions which covers some prospective regions which could be probed by FASER2 experiment in future.

Analyzing different interactions we find that some prospective bounds at NA64(eN)(eN), FASERν2\nu 2, JSNS2 could be probed in future. Some of the existing experimental limits from LEP, TEXONO, BaBaR(visible), dark photon searches at LHCb and CMS show stringent upper limits on gXg_{X} for the respective ZZ^{\prime} mass. JSNS2 bounds crosses the future sensitivities estimated from FASER, FASER2 and ILC-BD which could also be verified in future, however, their cross-overs depend on xHx_{H} which could be checked after the realistic experimental results will be available. Depending on U(1)XU(1)_{X} charge, we find that beam dump experiments like ν\nu-cal, E137, E141, NA64, E774, Orsay, CHARM, KEK, Nomad rules out the values of U(1)XU(1)_{X} coupling between 106gX0.0110^{-6}\leq g_{X}\leq 0.01 depending on the ZZ^{\prime} mass for MZ0.08M_{Z^{\prime}}\leq 0.08 GeV. Within the mass range of ZZ^{\prime}, we find that DUNE will provide a weaker bound from the beam dump scenario. We find that weaker limits obtained analyzing the g2g-2 data because in our model three generations of the fermions are equally coupled with the ZZ^{\prime}. We point out that limits from SND@@LHC, FASERν\nu are weak compared to the other scattering experiments below the ZZ-pole. Finally, from our analysis it has been found that scattering experiment could probe lighter ZZ^{\prime} between 0.020.02 GeV MZ0.2\leq M_{Z^{\prime}}\leq 0.2 GeV which could be simultaneously probed by proposed beam dump experiments involving FASER, FASER2 and ILC-BD. We find that in case of U(1)RU(1)_{R} scenario MUonE could provide a stringent bound for 0.020.02 GeV MZ0.175\leq M_{Z^{\prime}}\leq 0.175 GeV. On the other hand heavier ZZ^{\prime} above the ZZ-pole but MZ150M_{Z^{\prime}}\leq 150 GeV which could be probed by high energy colliders experiments in future for the cases we considered except xH=2x_{H}=-2. Hence we conclude that studying ZZ^{\prime} mediated interactions in addition to the SM processes, limits on general U(1)XU(1)_{X} couplings could be interesting to probe 𝒢SMU(1)X\mathcal{G}_{\rm SM}\otimes U(1)_{X} scenario in future.

Acknowledgments

We thank Takashi Shimomura and Yuichi Uesaka for useful advice. This work is supported by JSPS KAKENHI Grant Numbers JP21K20365 and JP23K13097 [KA], JP19K03860, JP19K03865 and JP23K03402 [OS], by the Fundamental Research Funds for the Central Universities [TN, JL], by the Natural Science Foundation of Sichuan Province under grant No. 2023NSFSC1329 and the National Natural Science Foundation of China under grant No. 11905149 [JL].

References