I review the Higgs sector of the U(1)B-L extension of the minimal supersymmetric standard model (MSSM). I will show that the gauge kinetic mixing plays a crucial role in the Higgs phenomenology. Two light bosons are present, a MSSM-like one and a B-L-like one, which mix at one loop solely due to the gauge mixing. After briefly looking at constraints from flavour observables, new decay channels involving right-handed (s)neutrinos are presented. Finally, how model features pertaining to the gauge extension affect the model phenomenology, concerning the existence of R-Parity-conserving minima at loop level and the Higgs-to-diphoton coupling, will be reviewed.
1. Introduction
The recently discovered Higgs boson is considered as the last missing piece of the standard model (SM) of particle physics. Nonetheless, several firm observations univocally call for its extension, mainly, but not limited to, the presence of dark matter, the neutrino masses and mixing pattern, the stability of the SM vacuum, and the hierarchy problem. Supersymmetry (SUSY) has long been considered as the most appealing framework to extend the SM. Its minimal realisations (MSSM and its constrained versions (for a review, see [1])) start however to feel considerable pressure to accommodate the recent findings, especially the measured Higgs mass of 125 GeV. Despite not in open contrast with the MSSM, the degree of fine tuning required to achieve it is more and more felt as unnatural. In order to alleviate this tension, nonminimal SUSY realisations can be considered. One can either extend the MSSM by the inclusion of extra singlets (e.g., NMSSM [2]) or by extending its gauge group. Concerning the latter, one of the simplest possibilities is to add an additional Abelian gauge group. I will focus here on the presence of U(1)B-L group which can be a result of an E8×E8 heterotic string theory (and hence M-theory) [3–5]. This model, the minimal R-Parity-conserving B-L supersymmetric standard model (BLSSM in short), was proposed in [6, 7] and neutrino masses are obtained via a type I seesaw mechanism. Furthermore, it could help to understand the origin of R-Parity and its possible spontaneous violation in supersymmetric models [6–8] as well as the mechanism of leptogenesis [9, 10].
It was early pointed out that the presence of two Abelian gauge groups in this model gives rise to kinetic mixing terms of the form(1)-χabF^a,μνF^μνb,a≠b,which are allowed by gauge and Lorentz invariance [11], as F^a,μν and F^b,μν are gauge-invariant quantities by themselves; see, for example, [12]. Even if these terms are absent at tree level at a particular scale, they will in general be generated by RGE effects [13, 14]. These terms can have a sizable effect on the mass spectrum of this model, as studied in detail in [15], and on the dark matter, where several scenarios would not work if kinetic mixing is neglected, as thoroughly investigated in [16]. In this work, I will review the properties of the Higgs sector of the model. Two light states exist, a MSSM-like boson and a B-L-like boson. After reviewing the model, I will show that a large portion of parameter space exists where the SM-like Higgs boson has a mass compatible with its measure, both in a “normal” (MH2>MH1=125 GeV) and in an “inverted” hierarchy (MH1<MH2=125 GeV), also in agreement with bounds from low energy observables and dark matter relic abundance. The phenomenological properties of the two lightest Higgs bosons will be systematically investigated, where once again the gauge mixing is shown to be fundamental. The presence of extra D-terms arising from the new U(1)B-L sector, as compared to models based on the SM gauge symmetry, has a large impact on the model phenomenology. They affect both the vacuum structure of the model and the Higgs sector, in particular enhancing the Higgs-to-diphoton coupling. Both of these issues will be reviewed here, although the latter is disfavoured by recent data [17], to show model features beyond the MSSM.
2. The Model
For a detailed discussion of the masses of all particles as well as of the corresponding one-loop corrections, we refer to [15]. Attention will be paid to the main aspects of the U(1) kinetic mixing since it has important consequence for the scalar sector. For the numerical investigations that will be shown, we used the SPheno version [22, 23] created with SARAH [24–28] for the BLSSM. For the standardised model definitions, see [29], while for a review of the model implementation in SARAH, see [30]. This spectrum calculator performs a two-loop RGE evaluation and calculates the mass spectrum at one loop. In addition, it calculates the decay widths and branching ratios (BRs) of all SUSY and Higgs particles as well as low energy observables like (g-2)μ. We will discuss the most constrained scenario with a universal scalar mass m0, a universal gaugino mass M1/2, and trilinear soft-breaking couplings proportional to the superpotential coupling (Ti=A0Yi) at the GUT scale. Other input parameters are tanβ, tanβ′, MZ′, Yx, and Yν. They will be defined in the following section. The numerical study here presented has been performed by randomly scanning over the independent input parameters above described via the SSP toolbox [31], while low energy observables such as BR(μ→eγ) and BR(μ→3e) have been evaluated with the FlavourKit package [32]. Furthermore, during the scans, all points have been checked with HiggsBounds-4.1.1 [33–36], both in the “normal” hierarchy and in the “inverted” hierarchy case.
2.1. Particle Content and Superpotential
The model consists of three generations of matter particles including right-handed neutrinos which can, for example, be embedded in SO(10) 16-plets. Moreover, below the GUT scale, the usual MSSM Higgs doublets are present as well as two fields η and η¯ responsible for the breaking of U(1)B-L. The η field is also responsible for generating a Majorana mass term for the right-handed neutrinos and thus we interpret its B-L charge as its lepton number. The same goes for η¯, and we call these fields bileptons since they carry twice the lepton number of (anti)neutrinos. The quantum numbers of the chiral superfields with respect to U(1)Y×SU(2)L×SU(3)C×U(1)B-L are summarised in Table 1.
Chiral superfields and their quantum numbers under GSM⊗U(1)B-L, where GSM=(U(1)Y⊗SU(2)L⊗SU(3)C).
Superfield
Spin-0
Spin-1/2
Generations
GSM⊗U(1)B-L
Q^
Q~
Q
3
(1/6,2,3,1/6)
d^c
d~c
dc
3
(1/3,1,3¯,-1/6)
u^c
u~c
uc
3
(-2/3,1,3¯,-1/6)
L^
L~
L
3
(-1/2,2,1,-1/2)
e^c
e~c
ec
3
(1,1,1,1/2)
ν^c
ν~c
νc
3
(0,1,1,1/2)
H^d
Hd
H~d
1
(-1/2,2,1,0)
H^u
Hu
H~u
1
(1/2,2,1,0)
η^
η
η~
1
(0,1,1,-1)
η-^
η-
η-~
1
(0,1,1,1)
The superpotential is given by(2)W=Yuiju^icQ^jH^u-Ydijd^icQ^jH^d-Yeije^icL^jH^d+μH^uH^d+Yνijν^icL^jH^u-μ′η^η¯^+Yxijν^icη^ν^jcand we have the additional soft SUSY-breaking terms:(3)LSB=LMSSM-λB~λB~′MBB′-12λB~′λB~′MB′-mη2η2-mη¯2η¯2-mνc,ij2ν~ic∗ν~jc-ηη¯Bμ′+TνijHuν~icL~j+Txijην~icν~jc,where i, j are generation indices. Without loss of generality, one can take Bμ and Bμ′ to be real. The extended gauge group breaks to SU(3)C⊗U(1)em as the Higgs fields and bileptons receive vacuum expectation values (vevs):(4)Hd0=12σd+vd+iϕd,Hu0=12σu+vu+iϕuη=12ση+vη+iϕη,η¯=12ση¯+vη¯+iϕη¯.We define tanβ′=vη/vη¯ in analogy to the ratio of MSSM vevs (tanβ=vu/vd).
2.2. Gauge Kinetic Mixing
As already mentioned in the Introduction, the presence of two Abelian gauge groups in combination with the given particle content gives rise to a new effect absent in any model with just one Abelian gauge group: gauge kinetic mixing. This can be seen most easily by inspecting the matrix of the anomalous dimension, which for our model at one loop reads(5)γ=116π23356256259,with typical GUT normalisation of the two Abelian gauge groups, that is, 3/5 for U1Y and 3/2 for U(1)B-L [7]. Therefore, even if at the GUT scale the U(1) kinetic mixing terms are zero, they are induced via RGE evaluation at lower scales. It turns out that it is more convenient to work with noncanonical covariant derivatives rather than with off-diagonal field-strength tensors as in (1). However, both approaches are equivalent [37]. Therefore, in the following, we consider covariant derivatives of the form Dμ=∂μ-iQϕTGA, where Qϕ is a vector containing the charges of the field ϕ with respect to the two Abelian gauge groups, G is the gauge coupling matrix(6)G=gYYgYBgBYgBB,and A contains the gauge bosons A=(AμY,AμB)T.
As long as the two Abelian gauge groups are unbroken, we have still the freedom to perform a change of basis by means of a suitable rotation. A convenient choice is the basis where gBY=0, since in this case only the Higgs doublets contribute to the gauge boson mass matrix of the SU(2)L⊗U(1)Y sector, while the impact of η and η¯ is only in the off-diagonal elements. Therefore, we choose the following basis at the electroweak scale [38]:(7)gYY′=gYYgBB-gYBgBYgBB2+gBY2=g1,gBB′=gBB2+gBY2=gBL,gYB′=gYBgBB+gBYgYYgBB2+gBY2=g¯,gBY′=0.
When unification at some large scale (~2·1016 GeV) is imposed, that is, g1GUT=g2GUT=gBLGUT and gYB′(GUT)=gBY′(GUT)=0, at SUSY scale, we get [15](8)gBL=0.548,g¯≃-0.147.
2.3. Tadpole Equations
The minimisation of the scalar potential is here described in the so-called tadpole method. We can solve the tree-level tadpole equations arising from the minimum conditions of the vacuum with respect to μ, Bμ, μ′, and Bμ′. Using vx2=vη2+vη¯2 and v2=vd2+vu2, we obtain(9)μ2=182g¯gBLvx2cos2β′-4mHd2+4mHu2·sec2β-4mHd2+mHu2-g12+g¯2+g22v2,(10)Bμ=-18-2g¯gBLvx2cos2β′+4mHd2-4mHu2+g12+g¯2+g22v2cos2βtan2β,(11)μ′2=14-2gBL2vx2+mη2+mη¯2+2mη2-2mη¯2+g¯gBLv2cos2βsec2β′,(12)Bμ′=14-2gBL2vx2cos2β′+2mη2-2mη¯2+g¯gBLv2·cos2βtan2β′,MZ′≃gBLvx, and, thus, we find an approximate relation between MZ′ and μ′(13)MZ′2≃-2μ′2+4mη¯2-mη2tan2β′-v2g¯gBLcosβ1+tanβ′2tan2β′-1.For the numerical results, the one-loop corrected equations are used, which lead to a shift of the solutions in (9)–(12).
2.4. The Scalar Sector
In this model, 2 MSSM complex doublets and 2 bilepton complex singlets are present, yielding 4CP-even, 2CP-odd, and 2 charged physical scalars.
Concerning the CP-even scalars, the MSSM and bilepton sectors are almost decoupled, mixing exclusively due to the gauge kinetic mixing. In first approximation, the mass matrix is block-diagonal and has mass eigenstates that mimic the MSSM case. In practice, it turns out that only two Higgs bosons are light (hereafter called H1 and H2, one per sector), while the other two are very heavy (above the TeV scale). The lightest scalars are well defined states, being either almost exclusively doublet-like or bilepton-like. It is worth stressing that their mixing is small (see Figure 4) and solely due to the gauge kinetic mixing (see also [39]).
Concerning the physical pseudoscalars A0 and Aη0, their masses are given by(14)mA02=2Bμsin2β,mAη02=2Bμ′sin2β′.For completeness, we note that the mass of charged Higgs boson reads as in the MSSM as(15)mH+2=Bμtanβ+cotβ+mW2.
In this model, the CP-odd and charged Higgses are typically very heavy. In (10), we see that, compared to the MSSM, there is a nonnegligible contribution from the gauge kinetic mixing. LHC searches limit tanβ′<1.5 and vx≳7TeV, since [40, 41](16)MZ′≳3.5TeVat 95% C.L. Notice that recent reanalysis of LEP precision data also constrains vx≳7TeV at 99% C.L. [42]. A consequence of this strong constraint in the BLSSM is that the first terms in (10)–(12) can be large, pushing for CP-odd and charged Higgs masses in the TeV range.
The very large bound on the Z′ mass is in contrast with the non-SUSY version of the model, where the gauge couplings are free parameters and can be much smaller, hence yielding lower mass bounds. The latter need to be evaluated as a function of both gauge couplings [43].
Next, we describe the sneutrino sector, which shows two distinct features compared to the MSSM. Firstly, it gets enlarged by the superpartners of the right-handed neutrinos. Secondly, even more drastically, a splitting between the real and imaginary parts of each sneutrino occurs resulting in twelve states: six scalar sneutrinos and six pseudoscalar ones [44, 45]. The origin of this splitting is the Yxijν^icη^ν^jc term in the superpotential (see (2)), which is ΔL=2 operator after the breaking of U(1)B-L. In the case of complex trilinear couplings or μ-terms, a mixing between the scalar and pseudoscalar particles occurs, resulting in 12 mixed states and consequently in a 12×12 mass matrix.
To gain some feeling for the behaviour of the sneutrino masses, we can consider a simplified setup: neglecting kinetic mixing as well as left-right mixing, the masses of the R-sneutrinos at the SUSY scale can be expressed as(17)mν~S2≃mνc2+MZ′214cos2β′+2Yx2gBL2sinβ′2+MZ′2YxgBLAxsinβ′-μ′cosβ′,mν~P2≃mνc2+MZ′214cos2β′+2Yx2gBL2sinβ′2-MZ′2YxgBLAxsinβ′-μ′cosβ′.In addition, we treat the parameters Ax, mνc2, MZ′, μ′, Yx, and tanβ′ as independent. The different effects on the sneutrino masses can easily be understood by inspecting (17). The first two terms give always a positive contribution whereas the third one gives a contribution that can be potentially large which differs in sign between the scalar and pseudoscalar states, therefore inducing a large mass splitting between the states. Further, this contribution can either be positive or negative depending on the sign of Axsinβ′-μ′cosβ′. For example, choosing Yx and μ′ positive, one finds that the CP-even (CP-odd) sneutrino is the lightest one for Ax<0 (Ax>0). This is pictorially shown in Figure 1, as a function of the GUT-scale input parameter A0, for a choice of the other parameters. One notices that the CP-even (CP-odd) sneutrino is the lightest one when the 125 GeV Higgs boson is predominantly H1 (H2). It is worth pointing out here that, as will be described in the following section, when MH1=125 GeV, the next-to-lightest Higgs boson can decay into pairs of CP-even sneutrinos, but not into similar channel with CP-odd sneutrinos. Being H2 predominantly a bilepton field, when this decay is open, it saturates its BRs; see Figure 3. Regarding the decay into CP-odd sneutrinos, this channel is accessible (i.e., ν~P is light enough) only in the region where H2 is the SM-like Higgs boson, that is, mainly coming from the doublets. In this case, however, this decay channel is mitigated by the small scalar mixing and is not overwhelming (unlike for H1, now mainly from the bileptons).
Masses of CP-even (ν~S, cyan) and CP-odd (ν~P, red) R-sneutrinos as a function of A0. For comparison, also the masses of the lightest (H1, black) and next-to-lightest (H2, blue) Higgs bosons are shown. Configurations when MH1=125GeV are shown in green.
Depending on the parameters, either type of sneutrinos can get very light. For the LSP, it can be a suitable dark matter candidate [16] and yield extra fully invisible decay channels to the Higgs bosons, thereby increasing their invisible widths. In the case of the decay into the CP-odd sneutrino, since this can happen mainly for the SM-like Higgs boson, one should account for the constraints on the former [17]. Eventually, the R-sneutrinos could also get tachyonic or develop dangerous R-Parity-violating vevs. While the first possibility is taken into account in our numerical evaluation by SPheno, and such points are excluded from our scans, the second case will be reviewed in the following subsection.
The last important sector for considerations that will follow is the one of the charged sleptons. See [46] for further details. New SUSY breaking D-term contributions to the masses appear, which can be parametrised as a function of the Z′ mass and of tanβ′ as(18)QB-L2MZ′tan2β′-11+tan2β′.Their impact is larger for the sleptons than for the squarks by a factor of 3 due to the different B-L charges (QB-L). It is possible to vary the stau mass by ±O(100) GeV with respect to the MSSM case while keeping the impact on the squarks under control. Having different sfermion masses in the BLSSM as compared to the MSSM has a net impact on the Higgs phenomenology, in particular in enhancing the hγγ coupling while keeping unaltered the SM-like Higgs coupling to gluons. As described at the end of this review, the new D-terms coming from the B-L sector can further reduce the stau mass entering in the hγγ effective interaction (while ensuring a pole mass of ~250GeV, compatible with exclusions) (with pole mass we denote the one-loop corrected mass at Q=MSUSY=t~1t~2, while in the loop, leading to the effective hγγ coupling, the running DR¯ tree-level mass at Q=mh enters, being h the SM-like Higgs boson; i.e., mh=125GeV) leading this mechanism to work also in the constrained version of the model. This mechanism has been recently reanalysed also in [47] in the very same model.
2.5. The Issue of R-Parity Conservation
We have encountered so far several neutral scalar fields which could develop vev, beside the Higgs bosons. If vevs of fields charged under QCD and electromagnetism are forbidden because the latter are good symmetries, R-sneutrino vevs, which are not by themselves problematic, would unavoidably break R-Parity. The issue of conserving R-Parity is of fundamental importance, since this is a built-in symmetry in our model where B-L is gauged. We will therefore restrain ourselves to parameter configurations where the global minimum is R-Parity conserving.
When all neutral scalar fields are allowed to get vev, it is not trivial even at the tree level to find which is the deeper global minimum and whether it is of a “good” type, here defined as having the correct broken symmetries and being R-Parity conserving. One possible way to study this issue is to start from a simplified set of input parameters yielding a correct tree-level global minimum when only the Higgs fields get vev. and then look for the true global minimum when all other neutral fields (mainly R-sneutrinos) acquire vev, both at the tree level and at loop level. See [48] for further details.
At the tree level there seems to exist regions where the BLSSM has a stable, R-Parity-conserving global minimum with the correct broken and unbroken gauge groups. For this to happen one needs the R-sneutrino Yukawa coupling Yx to be not so large and the trilinear parameter A0 to be not large compared to the soft scalar mass m0, as, intuitively, large Yx and A0 can lead to large negative contributions to the potential energy for large values of vx, as well as reducing the effective R-sneutrino masses, as described above and clear from Figure 1.
It turns out that when loop corrections are taken into account, few points all over such regions of parameters exist where R-Parity is not preserved anymore, or where SU(2)L or U(1)B-L is unbroken. This is apparently due to a very finely tuned breaking of SU(2)L and U(1)B-L which often does not survive loop corrections. The reason for this is that, besides the known large contributions of third generation (s)fermions, the additional new particles of the B-L sector also play an important role. As previously described for the charged sleptons sector, new SUSY breaking D-term contributions to the masses appear; see (18). Since, as shown in (16), the experimental bounds require MZ′ to be in the multi-TeV range, these contributions can be much larger than in the MSSM sector, resulting in the observed importance of the corresponding loop contributions. Furthermore, these contributions are also responsible for the restoration of U(1)B-L at the one-loop level.
Ultimately, overall safe regions of parameters cannot be found where the correct vacuum structure can be ensured. At the same time, if naive trends can be spotted for bad points to appear, these have nonetheless to be checked case by case due to the highly nontrivial scalar potential, and it might be possible that neighbour configurations still hold a valid global minimum. We will not check the validity of our scans from the vacuum point of view in the following, being confident that if any point is ruled out, a neighbour one yielding a very similar phenomenology can be found, which is allowed.
3. A Quick Look at Flavour Observables
Before moving to the Higgs phenomenology, we briefly show the impact on the BLSSM model when considering the constraints arising from low energy observables. For a review of the observables as well as for the impact on general SUSY models encompassing a seesaw mechanism, see [49, 50].
We consider here only the two most constraining ones, BR(μ→eγ) and BR(μ→3e). The present exclusions are BR(μ→eγ) <5.7·10-13 [18] and BR(μ→3e) <1·10-12 [19]. In Figure 2 we plot these branching ratios as a function of the mass of the lightest (in black) and next-to-lightest (in red) SM-like neutrino, which display some pattern for evading the bounds. In particular, they are required to be rather light, below 0.5eV, while the model, due to the limited scans here performed, seems to prefer configurations with neutrinos heavier than 0.01 eV, hence the preferred region in between. Lighter mass values are nonetheless also allowed.
(a) BR(μ→eγ) and (b) BR(μ→3e) as a function of the light neutrino masses in GeV (black: ν1, red: ν2). The blue horizontal lines represent the actual experimental limits, from [18] and [19], respectively. The parameters have been chosen as m0∈[0.4,2] TeV, M1/2∈[1.0,2.0] TeV, tanβ∈[5,40], A0∈[-4.0,4.0] TeV, tanβ′∈[1.05,1.15], MZ′∈[2.5,3.5] TeV, Yx∈1⋅[0.002,0.4], and Yν∈1⋅[0.05,5]×10-6.
Branching ratios for H2 with MH2>MH1=125GeV. The CP-even sneutrino channel (brown) is superimposed.
Mixing between Higgs boson mass eigenstates (blue-orange: H1, cyan-red: H2) and scalar doublet fields, as a function of MH2. ZH[i,j] is the scalar mixing matrix. Orange/red points are the subset corresponding to BR(H2→ν~Sν~S)>90%.
For convenience, the impact of satisfying the earlier bounds will be shown only in the inverted hierarchy case, due to the smaller density of configurations therein. Instead, points not allowed in the normal hierarchy case are automatically dropped.
Regarding the long-lasting (g-2)μ discrepancy, in the setup investigated here charginos and charged Higgses are too heavy, same for the Z′ boson, while the neutralino and sneutrino are too weakly coupled, to give a significant enhancement over the SM prediction.
4. Higgs Phenomenology
We review here the phenomenology of the Higgs sector, showing a first survey of its phenomenological features. First, results when normal hierarchy is imposed are presented. Then, we will show that the inverted hierarchy is also possible on a large portion of the parameter space. Without aim for completeness, the results are here presented as the starting point for a more thorough investigation. Finally, how model features pertaining to the extended gauge sector impinge on the Higgs phenomenology and in particular how the Higgs-to-diphoton branching ratio can be easily enhanced in this model, despite the experimental data now converging to a more SM-like behaviour than in the recent past, are described.
4.1. Normal Hierarchy
In this subsection we discuss the normal hierarchy case, with the lightest Higgs boson being the SM-like one (i.e., predominantly from the doublets), and a heavier Higgs boson predominantly from the bilepton fields (those carrying B-L number and responsible for the U(1)B-L spontaneous breaking). Their mixing is going to be small and solely due to the kinetic mixing.
In Figure 3 we first inspect the heavy Higgs boson branching ratios. Besides the standard decay modes, the decay into a pair of SM Higgs bosons exists, as well as two new characteristic channels of this model, comprising right-handed (s)neutrinos:
H2→H1H1. Its BR can be up to 40% before the top quark threshold and around 30% afterwards.
H2→νhνh. A similar decay channel exists for the Z′ boson. The BRs are O(10)%, up to 20% depending on the heavy Higgs and neutrino masses.
H2→ν~Sν~S, where, ν~S is the CP-even sneutrino and the LSP, hence providing fully invisible decays of the heavy Higgs. If kinematically open, it saturates the Higgs BRs. Notice that only points with very light CP-even sneutrinos are shown, possible only for very large and negative A0 (see Figure 1).
While the first two channels exist also in the non-SUSY version of the model (however, in the non-SUSY B-L model, the Higgs mixing angle is a free parameter, directly impacting these branching ratios) (see, e.g., [51]), the last one, involving the CP-even sneutrino, is truly new and rather intriguing. This is because the sneutrino is light and it can be a viable LSP candidate if its mass is smaller than H2, as in this case [16]. It however implies that the heavy Higgs is predominantly bilepton-like, with a light Higgs very much SM-like. This can be seen in Figure 4, where the points with large BR(H2→ν~Sν~S) (in red) have the lowest mixing between H2 and the SM scalar doublet fields, of the order of 0.1%. It immediately follows that this channel will have very small cross section at the LHC, when considering SM-like Higgs production mechanisms. This is true for all heavy Higgs masses MH2>140GeV. The 125GeV Higgs is well SM-like, with tiny reduction of its couplings to the SM particle content. On the other hand, the heavy Higgs is feebly mixed with the doublets, suppressing its interactions with the SM particles and hence its production cross section. This can be seen in Figure 5(a). Considering only the gluon-fusion production mechanism, and multiplying it by the relevant BR, we get the cross sections for the choice of channels displayed therein. The most constraining channels, H→WW→lνjj and H→WW→2l2ν, are also compared to the exclusions at the LHC for s=8TeV from [20] and [21], respectively. The H→ZZ channels are well below current exclusions, which are hence not shown.
Cross sections at s=8TeV for (a) the SM-like channels and (b) the new channels, as a function of the heavy Higgs mass. The solid lines above are the exclusion curves from [20, 21].
We see that all (starting from MH2>130GeV) the displayed configurations are allowed by the current searches (the exclusions shown by solid curves of the same color as the depicted channel). This is because of the suppression of the heavy Higgs boson cross sections due to the small scalar mixing.
In the lower plot the cross sections for the new channels are displayed. Those pertaining to model configurations for which the heavy Higgs boson decays to the CP-even sneutrino (LSP), yielding a fully invisible decay mode, are displayed in red. Contrary to all other cases, the production of the heavy Higgs for this channel is via vector boson fusion as searched for at the LHC [52]. Typical cross sections range between 0.1fb and 1fb. The H2→H1H1 channel is shown in blue and it can yield cross sections of 1÷10fb for 250<MH2<400GeV. Last is the H2→νhνh channel. It can be sizable only for very light H2 masses, ~10÷100fb for 140<MH2<160GeV, although the further decay chain of the heavy neutrinos has to be accounted for. The latter can give spectacular multileptonic final states of the heavy Higgs boson (4l2ν and 3l2jν) or high jet multiplicity ones (2l4j), via νh→l∓W± and νh→νZ in a 2 : 1 ratio (modulo threshold effects). Further, these decays are typically seesaw-suppressed and can therefore give rise to displaced vertices [53].
4.2. Inverted Hierarchy
In this subsection we discuss the inverted hierarchy case, where H2 is the SM-like boson and a lighter Higgs boson exists.
We start once again by presenting the BRs for the next-to-lightest Higgs boson in Figure 6. This time, however, this is the SM-like boson, hence predominantly from the doublets. It has the same new channels as the heavy Higgs in the normal hierarchy, the only difference being the CP-odd R-sneutrino instead of the CP-even one. This is simply because the inverted hierarchy can happen only for large positive A0 values, where only the CP-odd R-sneutrino can be light; see Figure 1. The configurations not allowed by the low energy observables or by HiggsBounds are displayed as gray points. We see that H2 may have sizable decays into pairs of the lighter Higgs bosons, yielding 4b-jets final states. This decay is still allowed with rates up to few percent. Further, rare decays into pairs of heavy neutrinos are also present, with BRs below the permil level. This channel can give rise to rare multilepton/jets decays for the SM-like Higgs boson, which are searched for at the LHC, even in combination with searches for displaced vertices [54]. The last available channel is the decay into pairs of CP-odd R-sneutrinos. Being the LSP, it will increase the invisible decay width and hence give larger-than-expected widths for the SM-like boson. Its rate is obviously constrained, and a precise evaluation of the allowed range is needed. It however goes beyond the scope of the present review and we postpone it to a future publication.
Branching ratios for the 125GeV Higgs boson (H2). The decay into heavy neutrinos is displayed with diamonds. All other decays are displayed with circles. Gray points are excluded by the low energy observables and by HiggsBounds. The decay into CP-odd sneutrinos is not shown.
Regarding the lightest Higgs boson (H1), this will obviously decay predominantly into pairs of b-jets, see Figure 7. Notice that due to its large bilepton fraction it can also decay into pairs of very light RH neutrinos, at sizable rates depending on the neutrino masses. As in Figure 6, the nonallowed configurations are displayed as gray points. We see that the pattern of decays is not affected by the inclusion of the constraints, in the sense that this channel stays viable. Once again, the latter will yield multilepton/jet final state, which will be very soft and hence very challenging for the LHC. However, also in this case displaced vertices may appear.
Same as in Figure 6 for the lightest Higgs boson (H1).
As in the previous section, we show in Figure 8 the mixing between the Higgs mass eigenstates and the doublet fields as a function of the light Higgs mass, to show that H2 is here rather SM-like. Once more, the gray points displayed here are excluded by the low energy observables and by HiggsBounds.
Mixing between scalar mass eigenstates and Higgs doublets (black: H1, red: H2) and scalar doublet fields, as a function of MH1. ZH[i,j] is the scalar mixing matrix. Gray points are excluded by the low energy observables and by HiggsBounds.
Finally, the production cross sections for the lightest Higgs boson can be evaluated. In Figure 9 we compare the direct production (for the main SM production mechanisms, gluon fusion and vector boson fusion) with the pair production via H2 decays only via gluon fusion, gg→H2→H1H1. When the latter channel is kinematically open, that is, 2MH1<125GeV, the lightest Higgs boson pair production has cross sections up to 1pb at the LHC at s=8TeV, and it can give rare 4b, 2b2V, or 4V (V=W,Z) decays of the SM-like Higgs boson. A thorough analysis of the phenomenology of the Higgs sector in the BLSSM for the upcoming LHC run 2, based on the first investigations shown here, will be performed soon.
Cross sections at s=8TeV for different production mechanisms. Gluon-fusion (in red) and vector-boson-fusion (in green) mechanisms are displayed only for MH1>50GeV for simplicity. Gray points are excluded by the low energy observables and by HiggsBounds.
4.3. Enhancement of the Diphoton Rate
A feature of gauge-extended models is that new SUSY-breaking D-terms arise, which give further contributions to the sparticle masses. In the case of the model under consideration, we showed discussing (18) that these terms can be large and that they bring larger corrections to sleptons than to squarks. We already discussed how the vacuum structure of the BLSSM is affected by this. Here, we discuss the impact of the new D-terms on the Higgs phenomenology, focusing on the Higgs-to-diphoton decay, despite being disfavoured by most recent data [17], as an illustrative case. See [46] for further details.
To start our discussion let us briefly review the partial decay width of the Higgs boson h into two photons within the MSSM and its singlet extensions. This can be written as (see, e.g., [55])(19)Γh→γγ=Gμα2mh31282π3∑fNcQf2ghffA1/2hτf+ghWWA1hτW+mW2ghH+H-2cW2mH±2A0hτH±+∑χi±2mWmχi±ghχi+χi-A1/2hτχi±+∑e~ighe~ie~ime~i2A0hτe~i+∑q~ighq~iq~imq~i23Qq~i2A0hτq~i2,corresponding to the contributions from charged SM fermions, W bosons, charged Higgs, charginos, charged sleptons, and squarks, respectively. The amplitudes Ai at the lowest order for the spin-1, spin-1/2, and spin-0 particle contributions can be found, for instance, in [55]. ghXX denotes the coupling between the Higgs boson and the particle in the loop and QX is its electric charge. In the SM, the largest contribution is given by the W-loop, while the top-loop leads to a small reduction of the decay rate. In the MSSM, it is possible to get large contributions due to sleptons and squarks, although it is difficult to realise such a scenario in a constrained model with universal sfermion masses [56–58]. In singlet or triplet extension of the MSSM also the chargino and charged Higgs can enhance the loop significantly [59, 60]. However, this is only possible for large singlet couplings which lead to a cut-off well below the GUT scale. In contrast, it is possible to enhance the diphoton ratio in the BLSSM due to light staus even in the case of universal boundary conditions at the GUT scale. We show this by calculating explicitly the contributions of the stau: (20)Aτ~=13∂detmτ~2∂logv≃-23·2mτ2Aτ-μtanβ2mE2+DRmL2+DL+mτ2μtanβ2Aτ-μtanβ.Here, DL and DR represent the D-term contributions of the left- and right-handed stau and we have neglected subleading contributions. Given that 2Aτ<μtanβ, for fixed values of the other parameters, DR and DL can be used to enhance the γγ rate by suppressing the denominator.
We turn now to a fully numerical analysis to demonstrate the mechanism to enhance the Higgs-to-diphoton rate as a feature of the model with an extended gauge sector. This is a result of reducing the stau mass at the Higgs mass scale via extra D-terms as shown discussing (18). We recall here that this mechanism leaves the stop mass and hence, as we will show, the Higgs-to-gluons effective coupling nearly unchanged. In Table 2 we have collected two possible scenarios that provide SM-like Higgs particle in the mass range preferred by LHC results displaying an enhanced diphoton rate. In the first point, the lightest CP-even scalar eigenstate is the SM-like Higgs boson while the light bilepton is roughly twice as heavy. In Figure 10 we show that all the features arise from the extended gauge sector: it is sufficient to change only tanβ′ to obtain an enhanced diphoton signal Rγγ1≡σ(gg→h1)·BR(h1→γγ)B-L/σ(gg→h1)·BR(h1→γγ)SM and the correct dark matter relic density while keeping the mass of the SM-like Higgs nearly unchanged. The dark matter candidate in this scenario is the lightest neutralino that is mostly a bileptino (the superpartner of the bileptons). The correct abundance for tanβ′≃1.156 is obtained due to a coannihilation with the light stau. In the second point, the SM-like Higgs is accompanied by a light scalar around 98GeV which couples weakly to the SM gauge bosons, compatibly with the LEP excess [61–63]. In this case, the LSP is a CP-odd sneutrino which annihilates very efficiently due to large Yx. This usually results in a small relic density. To get an abundance which is large enough to explain the dark matter relic, the mass of the sneutrino has to be tuned below mW [16]. This can be achieved by slightly increasing tanβ′ and by tuning the Majorana Yukawa couplings Yx, which tends to increase the SM-like Higgs mass for the given point. It is worth mentioning that a neutralino LSP with the correct relic density in the stau coannihilation region can also be found in this scenario. Notice that both points yield rates consistent with observations in the WW∗/ZZ∗ channels (measured at the LHC) (being chZZ~1), as well as an effective Higgs-to-gluon coupling close to 1.
The input parameter used: Point I: m0=673 GeV, M1/2=2220 GeV, A0=-1842 GeV, tanβ=42.2, tanβ′=1.1556, MZ′=2550 GeV, and Yx=1⋅0.42 (neutralino LSP); Point II: m0=742 GeV, M1/2=1572 GeV, A0=3277 GeV, tanβ=37.8, tanβ′=1.140, MZ′=2365 GeV, and Yx=diag(0.40,0.40,0.13) (CP-odd sneutrino LSP). cSVV denotes the coupling squared of the Higgs fields to vector bosons normalised to the SM values.
Point I
Point II
mh1 [GeV]
125.2
98.2
mh2 [GeV]
186.9
123.0
mτ~ [GeV]
267.0
237.3
Doublet fr. [%]
99.5
8.7
Bilepton fr. [%]
0.5
91.3
ch1gg
0.992
0.087
ch1ZZ
1.001
0.085
ch2gg
0.005
0.911
ch2ZZ
0.005
0.921
Γ(h1) [MeV]
4.13
0.22
Rγγ1
1.57
0.085
Rbb¯1
1.03
0.089
RWW∗1
0.98
0.05
Γ(h2) [MeV]
4.8
3.58
Rγγ2
0.005
1.79
Rbb¯2
0.006
0.95
RWW∗2
0.01
0.88
LSP mass [GeV]
253.9
82.9
Ωh2
0.10
10-2
(a) The mass of the SM-like Higgs (bottom (blue line)), of the stau (middle (black) line, where the dashed line represents a reference unchanged value), and of the lightest neutralino (top (red) line); (b) the diphoton branching ratio; (c) the neutralino relic density as a function of tanβ′. The other parameters have been chosen as m0=673GeV, M1/2=2220GeV, tanβ=42.2, A0=-1842.6, MZ′=2550GeV, and Yx=1⋅0.42.
5. Conclusions
In this review I described the U(1)B-L extension of the MSSM, focusing in particular on the scalar sector, described in detail. The fundamental role that the gauge kinetic mixing plays in this sector has been underlined.
The comparison to the most constraining low energy observables showed that a preferred region for the light neutrino masses exists to evade these bounds. Then, I presented a first systematic investigation of the phenomenology of the Higgs sector of this model, showing that both the normal hierarchy and the inverted hierarchy of the two lightest Higgs bosons are naturally possible in a large portion of the parameter space. Particular attention has been devoted to analysis of the new decay channels comprising both the CP-even and CP-odd R-sneutrinos, which are a peculiarity of the BLSSM. Based on these first findings, a thorough analysis of the Higgs sector in the BLSSM at the upcoming LHC run 2 will be soon prepared. The fit of the SM-like Higgs boson to the LHC data will also be performed with HiggsSignals [64].
Finally, I described how in the BLSSM model (and in general in gauge-extended MSSM models) the Higgs-to-diphoton decay can be easily enhanced. Despite being disfavoured by most recent data, this feature is a consequence of the potentially large new SUSY-breaking D-terms arising from the B-L sector. At the same time these terms affect also the vacuum structure of the model, where naive R-Parity-conserving configurations at the tree level could develop deeper R-Parity-violating global minima, or partially restore the SU(2)L×U(1)B-L symmetry at one loop. It is however possible to still find R-Parity-conserving global minima on the whole parameter space, which can either accommodate an enhancement of the Higgs-to-diphoton decay or fit the most recent Higgs data.
Conflict of Interests
The author declares that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
The author would like to thank S. Moretti and C. H. Shepherd-Themistocleous for helpful discussions in the early stages of this work. He is also really grateful to all his collaborators, and in particular to Florian Staub. He further acknowledges support from the Theorie-LHC France initiative of the CNRS/IN2P3 and from the French ANR 12 JS05 002 01 BATS@LHC.
EllisJ.LuoF.OliveK. A.SandickP.The Higgs mass beyond the CMSSM201373, article 240310.1140/epjc/s10052-013-2403-0EllwangerU.HugonieC.TeixeiraA. M.The next-to-minimal supersymmetric standard model20104961-217710.1016/j.physrep.2010.07.001MR27331162-s2.0-77957883789BuchmüllerW.HamaguchiK.LebedevO.RatzM.Supersymmetric Standard Model from the heterotic string (II)20077851-214920910.1016/j.nuclphysb.2007.06.028MR2357228AmbrosoM.OvrutB. A.THE B-L/electroweak hierarchy in smooth heterotic compactifications20102513263110.1142/S0217751X10049207AmbrosoM.OvrutB. A.The mass spectra, hierarchy and cosmology of B-L MSSM heterotic compactifications20102691569162710.1142/S0217751X11052943KhalilS.MasieroA.Radiative B-L symmetry breaking in supersymmetric models2008665537437710.1016/j.physletb.2008.06.063Fileviez PerezP.SpinnerS.Fate of R parity2011833703500410.1103/physrevd.83.035004BargerV.Fileviez PerezP.SpinnerS.Minimal gauged U(1)B-L model with spontaneous R parity violation200910218180210.1103/physrevlett.102.181802PeltoJ.ViljaI.VirtanenH.Leptogenesis B-L in gauged supersymmetry with MSSM Higgs sector20118305500110.1103/physrevd.83.055001BabuK. S.MengY.TavartkiladzeZ.New ways to leptogenesis with gauged B−L symmetry20096811374310.1016/j.physletb.2009.09.036HoldomB.Two U(1)'s and ϵ charge shifts1986166219619810.1016/0370-2693(86)91377-82-s2.0-0001285565BabuK. S.KoldaC.March-RussellJ.Implications of generalized Z-Z′ mixing199857116788679210.1103/physrevd.57.67882-s2.0-0000076036del AguilaF.CoughlanG. D.QuirósM.Gauge coupling renormalisation with several U(1) factors1988307363364810.1016/0550-3213(88)90266-02-s2.0-0000445336del AguilaF.GonzalezJ.QuirosM.Renormalization group analysis of extended electroweak models from the heterotic string1988307357163210.1016/0550-3213(88)90265-9O'LearyB.PorodW.StaubF.Mass spectrum of the minimal SUSY B-L model201220125, article 042BassoL.O'LearyB.PorodW.StaubF.Dark matter scenarios in the minimal SUSY B-L model201220129, article 5410.1007/JHEP09(2012)054KhachatryanV.SirunyanA. M.TumasyanA.Precise determination of the mass of the Higgs boson and tests of compatibility of its couplings with the standard model predictions using proton collisions at 7 and 8 TeV20147, article 21210.1140/epjc/s10052-015-3351-7AdamJ.BaiX.BaldiniA. M.New constraint on the existence of the μ+→e+γ decay201311020180110.1103/physrevlett.110.201801BellgardtU.OtterG.EichlerR.Search for the decay μ→ e+e+e−198829911610.1016/0550-3213(88)90462-2CMS CollaborationSearch for the standard model Higgs boson in the H to WW to lnujj decay channel in pp collisions at the LHC2012CMS-PAS-HIG-13-027Geneva, SwitzerlandCERNChatrchyanS.KhachatryanV.SirunyanA. M.Search for a standard-model-like Higgs boson with a mass in the range 145 to 1000 GeV at the LHC201373, article 246910.1140/epjc/s10052-013-2469-8PorodW.SPheno, a program for calculating supersymmetric spectra, SUSY particle decays and SUSY particle production at e+e- colliders2003153227531510.1016/S0010-4655(03)00222-4PorodW.StaubF.SPheno 3.1: extensions including flavour, CP-phases and models beyond the MSSM2012183112458246910.1016/j.cpc.2012.05.021StaubF.Sarahhttp://arxiv.org/abs/0806.0538StaubF.From superpotential to model files for FeynArts and CalcHep/CompHep201018161077108610.1016/j.cpc.2010.01.011StaubF.Automatic calculation of supersymmetric renormalization group equations and loop corrections2011182380883310.1016/j.cpc.2010.11.030ZBL1214.811682-s2.0-78650725626StaubF.SARAH 3.2: dirac gauginos, UFO output, and more201318471792180910.1016/j.cpc.2013.02.019StaubF.SARAH 4 : a tool for (not only SUSY) model builders201418561773179010.1016/j.cpc.2014.02.018BassoL.BelyaevA.ChowdhuryD.HirschM.KhalilS.MorettiS.O'LearyB.PorodW.StaubF.Proposal for generalised supersymmetry les Houches Accord for see-saw models and PDG numbering scheme2013184369871910.1016/j.cpc.2012.11.004ZBL1302.810072-s2.0-84872036170StaubF.Exploring new models in all detail with SARAHhttp://arxiv.org/abs/1503.04200StaubF.OhlT.PorodW.SpecknerC.A tool box for implementing supersymmetric models2012183102165220610.1016/j.cpc.2012.04.013PorodW.StaubF.VicenteA.A flavor kit for BSM models201474, article 299210.1140/epjc/s10052-014-2992-2BechtleP.BreinO.HeinemeyerS.WeigleinG.WilliamsK. E.HiggsBounds: confronting arbitrary Higgs sectors with exclusion bounds from LEP and the Tevatron2010181113816710.1016/j.cpc.2009.09.0032-s2.0-70449624838BechtleP.BreinO.HeinemeyerS.WeigleinG.WilliamsK. E.HiggsBounds 2.0.0: confronting neutral and charged Higgs sector predictions with exclusion bounds from LEP and the Tevatron2011182122605263110.1016/j.cpc.2011.07.0152-s2.0-80052322410BechtleP.BreinO.HeinemeyerS.StalO.StefaniakT.WeigleinG.WilliamsK.Recent developments in HiggsBounds and a preview of HiggsSignalsPoS CHARGED, http://arxiv.org/abs/1301.2345BechtleP.BreinO.HeinemeyerS.StalO.StefaniakT.WeigleinG.WilliamsK. E.HiggsBounds-4 : improved tests of extended Higgs sectors against exclusion bounds from LEP, the Tevatron and the LHC201474, article 269310.1140/epjc/s10052-013-2693-2FonsecaR. M.MalinskyM.PorodW.StaubF.Running soft parameters in SUSY models with multiple U(1) gauge factors20128541285310.1016/j.nuclphysb.2011.08.017ChankowskiP. H.PokorskiS.WagnerJ.Z′ and the Appelquist-Carrazzone decoupling200647118720510.1140/epjc/s2006-02537-3AbdallahW.KhalilS.MorettiS.Double Higgs peak in the minimal SUSY B-L model2015911601400110.1103/physrevd.91.014001AadG.AbbottB.AbdallahJ.Search for high-mass dilepton resonances in pp collisions at s=8 TeV with the ATLAS detector20149005200510.1103/physrevd.90.052005KhachatryanV.SirunyanA. M.TumasyanA.Search for physics beyond the standard model in dilepton mass spectra in proton-proton collisions at s-=8 TeV201520154, article 02510.1007/JHEP04(2015)025CacciapagliaG.CsákiC.MarandellaG.StrumiaA.The minimal set of electroweak precision parameters20067431203301110.1103/PhysRevD.74.033011BassoL.MimasuK.MorettiS.Non-exotic Z' signals in l+l-, bb and tt- final states at the LHC2012201211, article 06010.1007/JHEP11(2012)060HirschM.Klapdor-KleingrothausH. V.KovalenkoS. G.B-L-violating masses in softly broken supersymmetry19973983-431131410.1016/s0370-2693(97)00234-7GrossmanY.HaberH. E.Sneutrino mixing phenomena199778183438344110.1103/physrevlett.78.34382-s2.0-0000727385BassoL.StaubF.Enhancing h→γγ with staus in supersymmetric models with an extended gauge sector2013871601501110.1103/physrevd.87.015011HammadA.KhalilS.MorettiS.Higgs boson decays into γγ and Zγ in the MSSM and BLSSMhttp://arxiv.org/abs/1503.05408Camargo-MolinaJ. E.O'LearyB.PorodW.StaubF.Stability of R parity in supersymmetric models extended by U(1)B-L20138801503310.1103/PhysRevD.88.015033AbadaA.KraussM. E.PorodW.StaubF.VicenteA.WeilandC.Lepton flavor violation in low-scale seesaw models: SUSY and non-SUSY contributions2014201411, article 04810.1007/JHEP11(2014)048VicenteA.Lepton flavor violation beyond the MSSMhttp://arxiv.org/abs/1503.08622BassoL.MorettiS.PrunaG. M.Phenomenology of the minimal B-L extension of the standard model: the Higgs sector20118305501410.1103/PhysRevD.83.055014CMS CollaborationSearch for invisible decays of Higgs bosons in the vector boson fusion production mode2015CMS-PAS-HIG-14-038Geneva, SwitzerlandCERNhttp://cds.cern.ch/record/2007270BassoL.BelyaevA.MorettiS.Shepherd-ThemistocleousC. H.Phenomenology of the minimal B-L extension of the standard model: Z′ and neutrinos20098051405503010.1103/physrevd.80.055030BassoL.BelyaevA.FiaschiJ.MorettiS.TomalinI.ThomasM.In preparationDjouadiA.The anatomy of electroweak symmetry breaking: tome I: the Higgs boson in the Standard Model20084571–4121610.1016/j.physrep.2007.10.004CarenaM.GoriS.ShahN. R.WagnerC. E. M.A 125 GeV SM-like Higgs in the MSSM and the γγ rate201220123, article 01410.1007/JHEP03(2012)014EllwangerU.A Higgs boson near 125 GeV with enhanced di-photon signal in the NMSSM201220123, article 044BenbrikR.Gomez BockM.HeinemeyerS.StålO.WeigleinG.ZeuneL.Confronting the MSSM and the NMSSM with the discovery of a signal in the two photon channel at the LHC201272, article 217110.1140/epjc/s10052-012-2171-22-s2.0-84867011975Schmidt-HobergK.StaubF.Enhanced h→γγ rate in MSSM singlet extensions2012201210, article 19510.1007/JHEP10(2012)195DelgadoA.NardiniG.QuirosM.Large diphoton Higgs rates from supersymmetric triplets20128611711501010.1103/PhysRevD.86.115010ALEPH CollaborationDELPHI CollaborationL3 CollaborationOPAL CollaborationThe LEP Working Group for Higgs Boson SearchesSearch for the Standard Model Higgs boson at LEP20035656175030603310.1016/S0370-2693(03)00614-2BélangerG.EllwangerU.GunionJ. F.JiangY.KramlS.SchwarzJ. H.Higgs bosons at 98 and 125 GeV at LEP and the LHC201320131, article 06910.1007/JHEP01(2013)069DreesM.Supersymmetric explanation of the excess of Higgs-like events at the LHC and at LEP20128611501810.1103/physrevd.86.115018BechtleP.HeinemeyerS.StalO.StefaniakT.WeigleinG.HiggsSignals: confronting arbitrary Higgs sectors with measurements at the Tevatron and the LHC201474, article 271110.1140/epjc/s10052-013-2711-4