Two Higgs Doublet Models (2HDM)

• Two Higgs Doublet Models (2HDM) are extensions of the Standard Model featuring two SU(2) Higgs doublets that drive electroweak symmetry breaking and enrich the scalar spectrum.
• 2HDMs mitigate flavor-changing neutral currents through discrete symmetries and Yukawa alignment, providing controlled CP violation and robust experimental predictions.
• 2HDMs underpin various beyond-Standard-Model scenarios, including supersymmetry and inert models, with signatures like exotic Higgs decays and modified coupling measurements at colliders.

A Two-Higgs Doublet Model (2HDM) is an extension of the Standard Model in which the scalar sector consists of two SU(2)$_L$ Higgs doublets. This framework introduces a richer scalar spectrum, offers new sources of CP violation, and addresses several theoretical issues, such as the origin of mass hierarchies and the suppression of flavor-changing neutral currents (FCNCs). The 2HDM underpins the Higgs sectors in many beyond-the-Standard-Model scenarios, including supersymmetric models, and features a variety of realizations contingent on imposed symmetries, vacuum alignments, and Yukawa structures.

1. Model Structure and Scalar Spectrum

In a generic 2HDM, the two scalar doublets, $\Phi_1$ and $\Phi_2$, each with hypercharge $Y=+1$, participate in electroweak symmetry breaking. The most general renormalizable and CP-conserving scalar potential invariant under the Standard Model gauge group is: $$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$ After spontaneous symmetry breaking, the scalar spectrum consists of two neutral CP-even states ($h$ and $H$), one neutral CP-odd pseudoscalar ($A$), and a pair of charged Higgs bosons ($H^\pm$) (W. et al., 2012). The neutral CP-even states are admixtures of the real parts of the neutral components, with mixing governed by angle $\alpha$. The ratio of vevs $\Phi_1$0 parameterizes the alignment of the vacuum in the Higgs field space.

As a result, the physical masses are explicitly determined by the quartic couplings and $\Phi_1$1, with key relations: $\Phi_1$2 where $\Phi_1$3 (W. et al., 2012). The Goldstone bosons arising from the spontaneous breaking are absorbed as longitudinal components of the $\Phi_1$4 and $\Phi_1$5 bosons.

2. Yukawa Sector, FCNCs, and Symmetric Realizations

The general 2HDM produces tree-level FCNCs, originating from the two possible sources of fermion masses. This is phenomenologically disfavored, so various mechanisms are imposed:

• Discrete Symmetries (Z$\Phi_1$6-type): Imposing a $\Phi_1$7 symmetry (e.g., $\Phi_1$8, $\Phi_1$9) and restricting the coupling of fermions to individual doublets altogether enforces natural flavor conservation (Branco et al., 2011, Alves et al., 2018). This leads to the classic taxonomy:
  • Type I: All fermions couple to one doublet
  • Type II: Up-type quarks couple to $\Phi_2$0, down-type and leptons to $\Phi_2$1
  • Type X (lepton-specific) and Type Y (flipped): Different couplings for quark and lepton sectors
• Yukawa Alignment: The aligned 2HDM (A2HDM) postulates proportionality between Yukawa couplings of each doublet, parameterized by complex alignment parameters, ensuring the absence of tree-level FCNCs, but allowing for richer phenomenological consequences (Branco et al., 2011, Karan et al., 2024).
• Symmetry-Constrained and BGL Models: Abelian or non-abelian global symmetries further reduce the set of independent Yukawa couplings and control the structure of possible FCNCs, naturally tying them to the CKM matrix (Alves et al., 2018).
• Inert and Gauge-Symmetric Realizations: Z$\Phi_2$2-type "inert" models such as the inert doublet model (IDM) (0911.2457), and extensions such as G2HDM where the two Higgs doublets are embedded into a non-abelian gauge doublet (Huang et al., 2015), provide further variants that address FCNCs and give rise to distinctive phenomenology.

3. Vacuum Structure, Stability, and RG Evolution

The vacuum structure of the scalar potential is determined by minimizing with respect to both vevs. Stability at tree level requires the quartic couplings satisfy bounded-from-below conditions, e.g., $\Phi_2$3, $\Phi_2$4, $\Phi_2$5, and similar inequalities (W. et al., 2012).

Triviality and unitarity bounds are derived from RG evolution: requiring that none of the quartic couplings develops a Landau pole below a given scale sets upper constraints on their magnitudes, and that the scalar potential remains perturbative and stable up to a desired cutoff. The region of validity depends on initial values of couplings and the value of $\Phi_2$6, with extreme or semi-extreme cases yielding only short or moderate energy ranges before encountering nonperturbative behavior (W. et al., 2012).

In "asymptotically safe" 2HDMs, one can seek all quartic beta functions vanishing at the Planck scale, but only types II and Y in large $\Phi_2$7 admit such fixed point solutions—often at the expense of yielding a non-SM-like scalar spectrum or sacrificing absolute vacuum stability (Schuh, 2018).

4. CP Violation and Phenomenology

The 2HDM supports both explicit and spontaneous CP violation in the scalar sector via complex quartic couplings or vev alignments, respectively (Branco et al., 2011). CP violation can be quantified with basis-independent invariants (e.g., $\Phi_2$8 (Grzadkowski et al., 2010)). Even in the large $\Phi_2$9 limit—where mass degeneracies tend to quench CP violation—the invariants in 2HDM can reach $Y=+1$0, orders of magnitude above typical CKM-induced values in the SM.

The extended scalar sector leads to novel signatures:

• Charged Higgs bosons ($Y=+1$1) have distinctive phenomenology, with mass bounds driven by flavor-physics observables such as $Y=+1$2 (implying $Y=+1$3 GeV in type II, nearly independent of $Y=+1$4 (Arbey et al., 2017)).
• Neutral Higgses can decay via exotic channels (e.g., $Y=+1$5, $Y=+1$6) when mass splittings exceed $Y=+1$7, with rates peaking in the alignment limit where conventional $Y=+1$8 decay modes are suppressed (Kling et al., 2020).
• Multi-Higgs final states and triple-scalar self-couplings (e.g., $Y=+1$9, $$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$0) are accessible in both QCD and electroweak production, with electroweak contributions dominating in certain kinematical regimes (notably for $$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$1) (Enberg et al., 2017, Enberg et al., 2018, Enberg et al., 2018).
• Certain variants, such as the inert doublet model or G2HDM, allow dark matter candidates by virtue of a stable neutral scalar component protected by a discrete or gauge symmetry (0911.2457, Huang et al., 2015).

5. Experimental Constraints and LHC Phenomenology

The 2HDM parameter space is strongly constrained by a combination of:

• Direct searches for additional scalars at LEP, Tevatron, and the LHC (e.g., $$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$2), with typical lower mass limits for $$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$3 and extra neutral Higgses being model dependent, but often above 80–100 GeV (0911.2457, Arbey et al., 2017, Eberhardt, 2018).
• Precision measurements of the SM-like Higgs couplings, which drive approaches to the "alignment limit" ($$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$4), severely restricting deviations in the coupling structure (Eberhardt, 2018).
• Flavor physics observables, such as $$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$5, $$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$6-mixing, and leptonic $$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$7-decays, leading to exclusion of large swathes of the parameter space, especially in type II/Y realizations (Arbey et al., 2017, Eberhardt, 2018).
• Electroweak precision tests parameterized by oblique parameters ($$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$8, $$V = m_{11}^2 \Phi_1^\dagger \Phi_1 + m_{22}^2 \Phi_2^\dagger \Phi_2 - \left[m_{12}^2 \Phi_1^\dagger \Phi_2 + \text{h.c.}\right] + \frac{\lambda_1}{2} (\Phi_1^\dagger \Phi_1)^2 + \frac{\lambda_2}{2} (\Phi_2^\dagger \Phi_2)^2 + \lambda_3 (\Phi_1^\dagger \Phi_1)(\Phi_2^\dagger \Phi_2) + \lambda_4 (\Phi_1^\dagger \Phi_2)(\Phi_2^\dagger \Phi_1) + \frac{\lambda_5}{2} \left[\left(\Phi_1^\dagger \Phi_2\right)^2 + \text{h.c.}\right].$$9, $h$0), setting bounds on scalar mass splittings; correlated heavy state mass differences are often limited to $h$1–200 GeV (Eberhardt, 2018, Chang et al., 2015).

Distinct experimental signals include resonant and non-resonant Higgs pair production (sometimes yielding cross sections orders of magnitude above QCD pair production in special regions (Enberg et al., 2017, Enberg et al., 2018)), exotic decay cascades ($h$2, $h$3), and unusual signatures, such as multi-photon or multi-lepton final states in the fermiophobic or inert scenarios (Bernon et al., 2014, Enberg et al., 2018).

The current Higgs rate measurements and direct searches push models to a tightly constrained regime: for instance, global fits in (softly broken) $h$4 models place heavy Higgses above $h$5 GeV in type II models and restrict heavy mass splittings, with deviations from the alignment limit constrained to a few percent or less (Eberhardt, 2018).

6. Specialized Realizations: Inert, Hidden, Gauged, and Composite 2HDMs

Major 2HDM constructions with unique phenomenology include:

• Inert Doublet Model (Dark 2HDM): An exact $h$6 symmetry forbids $h$7 from acquiring a vev or Yukawa couplings; its neutral component is a dark matter candidate, with relic density and detection cross section predictions determined by scalar masses and couplings. LEP and LHC data imply $h$8 GeV, $h$9 GeV, with $H$0 heavier than 8 GeV (0911.2457).
• Hidden Light Higgs Scenarios: Switching the SM-like designation to the heavier CP-even Higgs ($H$1), with the lighter $H$2 "hidden," restricts the possible mass spectrum (e.g., $H$3 GeV) and requires soft $H$4 breaking below $H$5 to prevent large deviations in Higgs measurements (Chang et al., 2015).
• Gauged 2HDM (G2HDM): Both doublets form a doublet under a new $H$6 gauge group, inducing a scalar potential structure aligned to that symmetry and ensuring the stability of the inert neutral scalar as a dark matter candidate via gauge protection (Huang et al., 2015).
• Composite 2HDM: Two Higgs doublets emerge as pseudo-Goldstone bosons in a dilaton effective field theory, matching lattice SU(3) gauge theory results; custodial symmetry breaking, encoded by specific potential terms, yields scalar mass splittings contributing to the electroweak $H$7 parameter and addressing $H$8 deviations as in the recent CDF II result (Appelquist et al., 2022).

7. Future Directions and Theoretical Innovations

Numerous directions are under active exploration:

• Global Fits: Integrating up-to-date data in Bayesian frameworks (notably HEPfit) for both $H$9-symmetric and aligned 2HDMs, especially focusing on low-mass parameter regions and sensitivity to Yukawa alignment (Karan et al., 2024, Eberhardt, 2018).
• Dirac Algebra Formalism: A fully $A$0-covariant and IR-safe one-loop effective potential for general 2HDMs, exploiting field-space Dirac algebra and bilinears, yields a field-reparameterization–invariant approach to symmetry breaking and renormalization (Pilaftsis, 2024).
• Triple and Quartic Coupling Measurement: Direct access to the non-standard Higgs self-couplings (e.g., via LHC and $A$1 colliders), especially in multi-Higgs, exotic decay, or enhanced di-Higgs cross section regions (Enberg et al., 2018, Chang et al., 2015).
• Interplay of Theory and Experiment: The continuing refinement of parameter space against LHC and future collider data, flavor and precision observables, is driving both extensions (e.g., addition of dark matter, CP violation, inert or gauge-extended sectors) and innovation in analysis methods (e.g., advanced Monte Carlo sampling, boosted decision tree techniques) (Hanson et al., 2018, Enberg et al., 2018).

The 2HDM framework continues to serve as a principal benchmark in searches for new physics, with its theoretical landscape, phenomenological signatures, and experimental constraints thoroughly mapped but still supporting a wide array of realistic and testable scenarios.