On vacua and bounded masses in the general 2HDM

José M. Camacho, Carlos Miró, Miguel Nebot, Tomás Tobarra · 2026

Read On vacua and bounded masses in the general 2HDM by José M. Camacho, Carlos Miró, Miguel Nebot, Tomás Tobarra (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-02710.txt.

Introduction

Camacho and co-authors prove a mass bound for two-Higgs-doublet models with two vacua. Such models extend the Higgs sector with a second doublet. Their scalar potential can break electroweak symmetry at one or two local minima. With one minimum, a decoupling regime exists. All new scalars can then be heavy. The two-minimum case behaves differently.

For potentials with two local minima, all scalar masses are bounded. The bound holds when the dimensionless quartic couplings respect perturbativity limits. No decoupling escape remains in that branch. Model builders must therefore face light scalars or tuned couplings. The result sharpens the map of allowed two-Higgs-doublet theories. It also guides collider searches for the extra Higgs states.

Catalog record

CREATORJosé M. Camacho, Carlos Miró, Miguel Nebot, Tomás Tobarra
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE30 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.02710

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On vacua and bounded masses in the general 2HDM José M. Camacho, Carlos Miró, Miguel Nebot, Tomás Tobarra Abstract Two Higgs doublets models with a scalar potential that breaks electroweak symmetry spontaneously can have either one or two local minima. While potentials with one minimum can have a decoupling regime where all the new scalars are heavy, we show that, for potentials with two local minima, the masses of all the scalars are bounded if the dimensionless quartic couplings obey perturbativity constraints. I Introduction Knowing the masses of new particles arising in scenarios beyond the Standard Model (SM) is of great importance. Apart from direct or indirect experimental information, theoretical arguments have been put forward to constrain possible mass spectra 34 ; 33 ; 4 ; 8 ; 5 , including in particular perturbativity requirements as proposed by Lee, Quigg and Thacker 25 ; 24 ; 9 to set bounds on the Higgs mass. In extended scalar sectors inducing spontaneous symmetry breaking, these perturbativity requirements imply the presence of a Higgs-like scalar with electroweak-scale mass, set by the spontaneous symmetry breaking vacuum expectation value (vev) v_{\rm EW}\simeq 246 GeV 35 ; 23 ; 7 ; 10 . In multi-Higgs models 20 , starting with the minimal scenario with two doublets (2HDM) 26 ; 27 , these constraints have been explored, for example, in 17 ; 6 ; 28 ; 21 ; 11 ; 16 ; 22 . In some cases, not necessarily multi-Higgs scenarios, scalar sectors where an imposed symmetry yields a number of quadratic parameters in the potential not larger than the number of stationarity conditions, the quadratic parameters can be expressed in terms of bounded quartic parameters and vacuum expectation values, giving as a result that all the new scalars have bounded masses. One illustrative case is the 2HDM where the potential is CP invariant and the vacuum breaks CP spontaneously: the 3 real quadratic parameters in the potential are reexpressed in terms of dimensionless quartic parameters and vevs, and then perturbativity constraints on the quartic parameters force \mathcal{O}(v_{\rm EW}) bounds on all scalar masses 30 ; 31 ; 32 . With more than two doublets, a CP invariant potential and spontaneous CP violation (SCPV), there are free, unconstrained quadratic parameters. Unexpectedly, this freedom cannot produce masses much larger than the electroweak scale for all the new scalars, as discussed in 29 ; Camacho:2026hrc . One critical ingredient appears to be the following: in that context, if there is a CP violating candidate vacuum (a minimum of the potential), its CP-conjugate is also a candidate vacuum (with, in fact, the same depth). This motivates the present work: since the general 2HDM can have either one or two possible vacua 19 , is something similar at work? 2HDMs, or at least some of them, can have a decoupling regime for the new scalars 14 ; 13 ; the goal here is to provide a different perspective: what can be said, concerning the mass spectrum, depending on the fact that the scalar potential has either one or two local minima? The surprising answer is that in the case of a potential with two minima, the masses of all the scalars are bounded, they cannot be much larger than the electroweak scale, if perturbativity constraints are imposed on the quartic couplings. The manuscript is organized as follows. In section II we set notation and address some basic aspects. A detailed numerical analysis is presented in section III . In section IV we discuss the central results, and then present our conclusions. II The general 2HDM The most general scalar potential for two SU(2)_{L} doublets with hypercharge 1/2 , \Phi_{1}^{\phantom{\dagger}} and \Phi_{2}^{\phantom{\dagger}} , is \displaystyle\mathcal{V}(\Phi_{1}^{\phantom{\dagger}},\Phi_{2}^{\phantom{\dagger}}) \displaystyle=\mathcal{V}_{2}(\Phi_{1}^{\phantom{\dagger}},\Phi_{2}^{\phantom{\dagger}})+\mathcal{V}_{4}(\Phi_{1}^{\phantom{\dagger}},\Phi_{2}^{\phantom{\dagger}}), (1) \displaystyle\mathcal{V}_{2}(\Phi_{1}^{\phantom{\dagger}},\Phi_{2}^{\phantom{\dagger}}) \displaystyle=\mu^{2}_{1}\,\Phi_{1}^{\dagger}\Phi_{1}^{\phantom{\dagger}}+\mu^{2}_{2}\,\Phi_{2}^{\dagger}\Phi_{2}^{\phantom{\dagger}}+\mu^{2}_{12}\,\Phi_{1}^{\dagger}\Phi_{2}^{\phantom{\dagger}}+\mu^{2\,\ast}_{12}\,\Phi_{2}^{\dagger}\Phi_{1}^{\phantom{\dagger}}\,, \displaystyle\mathcal{V}_{4}(\Phi_{1}^{\phantom{\dagger}},\Phi_{2}^{\phantom{\dagger}}) \displaystyle=\lambda_{1}(\Phi_{1}^{\dagger}\Phi_{1}^{\phantom{\dagger}})^{2}+\lambda_{2}(\Phi_{2}^{\dagger}\Phi_{2}^{\phantom{\dagger}})^{2}+\lambda_{3}(\Phi_{1}^{\dagger}\Phi_{1}^{\phantom{\dagger}})(\Phi_{2}^{\dagger}\Phi_{2}^{\phantom{\dagger}})+\lambda_{4}(\Phi_{1}^{\dagger}\Phi_{2}^{\phantom{\dagger}})(\Phi_{2}^{\dagger}\Phi_{1}^{\phantom{\dagger}}) \displaystyle+\lambda_{5}(\Phi_{1}^{\dagger}\Phi_{2}^{\phantom{\dagger}})^{2}+\lambda_{5}^{\ast}(\Phi_{2}^{\dagger}\Phi_{1}^{\phantom{\dagger}})^{2} \displaystyle+(\lambda_{6}\,\Phi_{1}^{\dagger}\Phi_{2}^{\phantom{\dagger}}+\lambda_{6}^{\ast}\,\Phi_{2}^{\dagger}\Phi_{1}^{\phantom{\dagger}})(\Phi_{1}^{\dagger}\Phi_{1}^{\phantom{\dagger}})+(\lambda_{7}\,\Phi_{1}^{\dagger}\Phi_{2}^{\phantom{\dagger}}+\lambda_{7}^{\ast}\,\Phi_{2}^{\dagger}\Phi_{1}^{\phantom{\dagger}})(\Phi_{2}^{\dagger}\Phi_{2}^{\phantom{\dagger}})\,, where \mu^{2}_{1} , \mu^{2}_{2} , \lambda_{1} , \lambda_{2} , \lambda_{3} , \lambda_{4} are necessarily real, while \mu^{2}_{12} , \lambda_{5} , \lambda_{6} , \lambda_{7} are, in general, complex. Spontaneous symmetry breaking occurs for a minimum of the function \mathrm{V}(\langle\Phi_{1}^{\phantom{\dagger}}\rangle,\langle\Phi_{2}^{\phantom{\dagger}}\rangle)\equiv\mathcal{V}(\langle\Phi_{1}^{\phantom{\dagger}}\rangle,\langle\Phi_{2}^{\phantom{\dagger}}\rangle)\,, (2) with \langle\Phi_{a}^{\phantom{\dagger}}\rangle\neq\left(\begin{smallmatrix}0\\ 0\end{smallmatrix}\right) . Electroweak spontaneous symmetry breaking is assumed for vevs \langle\Phi_{a}^{\phantom{\dagger}}\rangle=\frac{v_{a}e^{i\theta_{a}}}{\sqrt{2}}\begin{pmatrix}0\\ 1\end{pmatrix},\qquad v_{a},\theta_{a}\in\mathbb{R}\,,\quad v_{a}\geq 0\,, (3) and the fields are expanded as \Phi_{a}^{\phantom{\dagger}}=\frac{e^{i\theta_{a}}}{\sqrt{2}}\begin{pmatrix}\sqrt{2}\mathrm{C}^{+}_{a}\\ v_{a}+\mathrm{R}_{a}+i\,\mathrm{I}_{a}\end{pmatrix}. (4) Only the phase difference \theta\equiv\theta_{2}-\theta_{1} (mod 2\pi ) is physical and can produce spontaneous CP violation. We use v\equiv\sqrt{v_{1}^{2}+v_{2}^{2}} ; for adequate electroweak symmetry breaking, one sets v=v_{\rm EW} . The 3 stationarity conditions, necessarily fulfilled at a minimum of the potential, are \displaystyle\partial_{v_{1}}\mathrm{V} \displaystyle=\mu^{2}_{1}v_{1}+\text{Re}\left(\mu^{2}_{12}e^{i\theta}\right)v_{2}+\lambda_{1}v_{1}^{3}+\frac{1}{2}(\lambda_{3}+\lambda_{4})v_{1}v_{2}^{2} (5) \displaystyle+\text{Re}\left(\lambda_{5}e^{i2\theta}\right)v_{1}v_{2}^{2}+\frac{3}{2}\text{Re}\left(\lambda_{6}e^{i\theta}\right)v_{1}^{2}v_{2}+\frac{1}{2}\text{Re}\left(\lambda_{7}e^{i\theta}\right)v_{2}^{3}=0\,, \displaystyle\partial_{v_{2}}\mathrm{V} \displaystyle=\mu^{2}_{2}v_{2}+\text{Re}\left(\mu^{2}_{12}e^{i\theta}\right)v_{1}+\lambda_{2}v_{2}^{3}+\frac{1}{2}(\lambda_{3}+\lambda_{4})v_{1}^{2}v_{2} (6) \displaystyle+\text{Re}\left(\lambda_{5}e^{i2\theta}\right)v_{1}^{2}v_{2}+\frac{1}{2}\text{Re}\left(\lambda_{6}e^{i\theta}\right)v_{1}^{3}+\frac{3}{2}\text{Re}\left(\lambda_{7}e^{i\theta}\right)v_{1}v_{2}^{2}=0\,, \displaystyle\partial_{\theta}\mathrm{V} \displaystyle=-\text{Im}\left(\mu^{2}_{12}e^{i\theta}\right)v_{1}v_{2}-\text{Im}\left(\lambda_{5}e^{i2\theta}\right)v_{1}^{2}v_{2}^{2}-\text{Im}\left((\lambda_{6}v_{1}^{2}+\lambda_{7}v_{2}^{2})e^{i\theta}\right)\frac{v_{1}v_{2}}{2}=0\,. (7) Notice that, with 4 quadratic parameters and 3 stationarity conditions, one cannot engineer a potential with the chosen vevs and all quadratic parameters fixed in terms of the vevs and other parameters: one typically chooses the vacuum configuration \{v_{1},v_{2}e^{i\theta}\} , and fixes \mu^{2}_{1} , \mu^{2}_{2} and \text{Re}\left(\mu^{2}_{12}\right) through eqs. 5 , 6 and 7 , while \text{Im}\left(\mu^{2}_{12}\right) is free. The mass terms for the charged {\vec{\mathbf{C}}}^{\dagger}=(\mathrm{C}^{+}_{1},\mathrm{C}^{+}_{2}) and neutral fields {\vec{\mathbf{N}}}^{T}=(\mathrm{R}_{1},\mathrm{R}_{2},\mathrm{I}_{1},\mathrm{I}_{2}) are -\mathscr{L}_{\mathrm{Mass}}={\vec{\mathbf{C}}}^{\dagger}\,{\rm M_{\pm}^{2}}\,{\vec{\mathbf{C}}}+\frac{1}{2}{\vec{\mathbf{N}}}^{T}\,{\rm M_{0}^{2}}\,{\vec{\mathbf{N}}}\,, (8) where {\rm M_{\pm}^{2}} is the 2\times 2 hermitian charged mass matrix and {\rm M_{0}^{2}} is the 4\times 4 real symmetric neutral mass matrix. {\rm M_{\pm}^{2}} and {\rm M_{0}^{2}} have null eigenvalues corresponding to the would-be Goldstone bosons, with eigenvectors \displaystyle\vec{c}_{G}=(v_{1},v_{2})^{T}\,,\qquad \displaystyle\rm{\displaystyle M_{\pm}^{2}}\,\vec{c}_{G}=\left(\partial_{v_{1}}\mathrm{V}+\frac{i}{v_{1}}\partial_{\theta}\mathrm{V},\ \partial_{v_{2}}\mathrm{V}-\frac{i}{v_{2}}\partial_{\theta}\mathrm{V}\right)^{T}=(0,0)^{T}\,, (9) \displaystyle\vec{n}_{G}=(0,0,v_{1},v_{2})^{T}\,,\qquad \displaystyle\rm{\displaystyle M_{0}^{2}}\,\vec{n}_{G}=\left(\frac{1}{v_{1}}\partial_{\theta}\mathrm{V},\ -\frac{1}{v_{2}}\partial_{\theta}\mathrm{V},\ \partial_{v_{1}}\mathrm{V},\ \partial_{v_{2}}\mathrm{V}\right)^{T}=(0,0,0,0)^{T}\,. {\rm M_{\pm}^{2}} is diagonalized with a 2\times 2 unitary matrix U , which gives the would-be Goldstone bosons G^{\pm} and the physical charged scalar \mathrm{H}^{\pm} : (\mathrm{C}^{-}_{1},\mathrm{C}^{-}_{2})^{T}=U\,(G^{-},\mathrm{H}^{-})^{T}\,,\qquad U^{\dagger}\,{\rm M_{\pm}^{2}}\,U=\text{Diag}(0,M_{\mathrm{H}^{\pm}}^{2})\,. (10) {\rm M_{0}^{2}} is diagonalized with a 4\times 4 real orthogonal matrix O , which gives the would-be Goldstone boson G^{0} and the physical neutral scalars. We assume that the lightest neutral scalar is “the Higgs” h (this is just a convenient choice). We have (\mathrm{R}_{1},\mathrm{R}_{2},\mathrm{I}_{1},\mathrm{I}_{2})^{T}=O\,(G^{0},h,\mathrm{H}_{1}^{0},\mathrm{H}_{2}^{0})^{T}\,,\qquad O^{T}\,{\rm M_{0}^{2}}\,O=\text{Diag}(0,m_{h}^{2},M_{\mathrm{H}_{1}^{0}}^{2},M_{\mathrm{H}_{2}^{0}}^{2})\,, (11) with m_{h}\leq M_{\mathrm{H}_{1}^{0}}\leq M_{\mathrm{H}_{2}^{0}} . The diagonalization of {\rm M_{\pm}^{2}} is straightforwardly achieved by rotating to “the Higgs basis” 3 with U=\frac{1}{v}\begin{pmatrix}v_{1}&-v_{2}\\ v_{2}&v_{1}\end{pmatrix}\,. (12) From the diagonalization, or computing directly \text{Tr}\left({\rm M_{\pm}^{2}}\right) , one obtains the mass of the charged scalar M_{\mathrm{H}^{\pm}}^{2}=\mu^{2}_{1}+\mu^{2}_{2}+\lambda_{1}v_{1}^{2}+\lambda_{2}v_{2}^{2}+\frac{1}{2}\lambda_{3}v^{2}+\text{Re}\left((\lambda_{6}+\lambda_{7})e^{i\theta}\right)v_{1}v_{2}\,, (13) expressed for simplicity in terms of \mu^{2}_{1}+\mu^{2}_{2} as the free quadratic parameter left; one can equivalently express M_{\mathrm{H}^{\pm}}^{2} , for example, in terms of \text{Im}\left(\mu^{2}_{12}\right) : M_{\mathrm{H}^{\pm}}^{2}=\frac{v^{2}}{2v_{1}v_{2}\sin\theta}\text{Im}\left(2\mu^{2}_{12}+(2\lambda_{5}-\lambda_{4})e^{i\theta}v_{1}v_{2}+\lambda_{6}v_{1}^{2}+\lambda_{7}v_{2}^{2}\right)\,. (14) For {\rm M_{0}^{2}} , the calculation of the eigenvalues is not straightforward, but it is sufficient for now to notice that, from \text{Tr}\left({\rm M_{0}^{2}}\right) , m_{h}^{2}+M_{\mathrm{H}_{1}^{0}}^{2}+M_{\mathrm{H}_{2}^{0}}^{2}=2(\mu^{2}_{1}+\mu^{2}_{2})+4(\lambda_{1}v_{1}^{2}+\lambda_{2}v_{2}^{2})+(\lambda_{3}+\lambda_{4})v^{2}+4\text{Re}\left((\lambda_{6}+\lambda_{7})e^{i\theta}\right)v_{1}v_{2}\,. (15) From eqs. 13 and 15 , it is clear that giving masses to the new scalars much larger than the electroweak scale v_{\rm EW} , requires \mu^{2}_{1}+\mu^{2}_{2}\gg v_{\rm EW}^{2} . The sole use of the stationarity conditions in eqs. 5 , 6 and 7 does not appear to bar that possibility, and thus a decoupling regime for (all) the new scalars is possible. A crucial comment is in order. We have mentioned the vacuum \{v_{1},v_{2}e^{i\theta}\} and have discussed accordingly the stationarity conditions, but it might happen that the scalar potential has not one but two local symmetry breaking minima, as analyzed in depth in 19 (see also 1 ; 2 ; 18 ; 12 ). If that is the case, then a second solution \{v_{1}^{\prime},v_{2}^{\prime}e^{i\theta^{\prime}}\} of eqs. 5 , 6 and 7 exists. Then, the possibility of having a decoupling regime might be jeopardized: since we would have a number of conditions to be fulfilled in excess of

Read the complete text