Coupled Quintom Model in Cosmology

• The coupled quintom model is a cosmological framework featuring a canonical scalar and a phantom scalar that interact to enable smooth crossing of the w=-1 barrier.
• It employs symmetry-based methods and varied coupling schemes—including potential, curvature, and teleparallel couplings—to derive integrable solutions and diverse expansion histories.
• Its implications include explaining dark energy dynamics, alleviating Hubble tension, and providing alternatives to ΛCDM with scenarios like de Sitter, bouncing, and matter-like phases.

A coupled quintom model is a cosmological construction in which the effective dark-energy equation of state crosses the cosmological-constant boundary $w=-1$, while the dark sector contains either two degrees of freedom with opposite-sign kinetic structure or a single effective component whose coupling to additional geometric variables removes the usual crossing obstruction. In the standard minimal realization, the model is a flat-FRW two-field system with one canonical scalar and one phantom scalar, coupled through a shared potential $V(\phi,\sigma)$; later generalizations replace or supplement this potential coupling by curvature, torsion, matter-trace, spin-torsion, gauge, or Nieh–Yan couplings (Aslam et al., 2013, Bahamonde et al., 2018, Kang et al., 31 Jan 2026).

1. Canonical two-field structure

In the minimal Einstein-gravity formulation, the coupled quintom sector is built from a quintessence field $\phi$ and a phantom field $\sigma$ with opposite-sign kinetic terms. A representative action used in the literature is

$$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$

and, in flat FRW, the corresponding point-like Lagrangian is

$$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$

The effective energy density and pressure are

$$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$

so that

$$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$

The scalar equations are

$$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$

with the sign difference in the force term reflecting the phantom nature of $\sigma$ (Aslam et al., 2013, Dutta et al., 2016).

In this minimal setting, the coupling is purely potential-based: there is no kinetic mixing and no nonminimal coupling to gravity. This point is explicit in the symmetry-based analyses of flat-FRW quintom cosmology, where the interaction is entirely encoded in the mixed dependence of $V(\phi,\sigma)$0 on both fields (Aslam et al., 2013). Anisotropic extensions preserve this distinction. In Bianchi I, Bianchi III, and Kantowski–Sachs backgrounds, the two fields still enter with diagonal kinetic terms, and the paper on inflation and isotropization distinguishes a multiplicative potential $V(\phi,\sigma)$1 from a collective mode $V(\phi,\sigma)$2; in both cases, the coupling is mediated by the common potential rather than by a direct kinetic term (Tajahmad, 2024).

2. Symmetry-selected interaction potentials

A major strand of coupled quintom research constrains $V(\phi,\sigma)$3 by demanding Lie or Noether symmetries of the minisuperspace dynamics. In the Noether gauge symmetry analysis of flat-FRW quintom cosmology, the main interacting family is

$V(\phi,\sigma)$4

This family includes the quadratic interaction potentials used phenomenologically in the quintom literature, but here they arise from symmetry rather than ad hoc choice. For constant gauge $V(\phi,\sigma)$5, the interacting family admits the nontrivial generator

$V(\phi,\sigma)$6

which mixes $V(\phi,\sigma)$7 and $V(\phi,\sigma)$8 and singles out a genuine internal symmetry of the coupled system (Aslam et al., 2013).

Related Lie and Noether analyses yield other symmetry-selected potentials. One paper finds a Lie-symmetry restriction

$V(\phi,\sigma)$9

while its Noether analysis gives the polynomial family

$\phi$0

The same work emphasizes that the Noether algebra is a subalgebra of the Lie algebra, and that the polynomial potential is organized by the hyperbolic combination $\phi$1, which is natural for a field space with one canonical and one phantom direction (Dutta et al., 2016).

The later reassessment of the Noether gauge symmetry program extends this structure further. It derives two explicitly nonlinear coupled families,

$\phi$2

and argues that several Noether symmetries and invariants reported earlier were incorrect. It also states that one previously proposed cosmological model was therefore not a feasible quintom model (Ali, 2015). This correction matters because the symmetry algebra is used not only to classify admissible interactions but also to justify conserved quantities and reduced dynamics.

3. Integrability, conserved quantities, and exact cosmological sectors

The symmetry approach is not limited to model selection; it also provides first integrals and exact solutions. In the Noether gauge symmetry treatment, the autonomous time-translation generator $\phi$3 corresponds to energy conservation, while additional generators supply nontrivial Noether invariants that reduce the effective order of the system. For the symmetry-derived interacting family, a specific numerical choice $\phi$4, $\phi$5, and $\phi$6 gives

$\phi$7

which was used for explicit numerical cosmology. The resulting dynamics exhibits accelerated expansion, $\phi$8CDM-like Hubble evolution, phantom-divide crossing, and stable attractor behavior in the sense stated in that analysis (Aslam et al., 2013).

Symmetry reduction also exposes exact integrable subsectors. In one Noether-invariant case, the scale factor takes either a de Sitter form,

$\phi$9

or a power law,

$\sigma$0

with the authors identifying these classes as de Sitter and power-law solutions (Aslam et al., 2013). In a separate Lie–Noether treatment, a point transformation in the augmented configuration space $\sigma$1 is chosen so that one variable becomes cyclic, producing compact first integrals for the Hamiltonian and Noether charge and allowing exact solution branches to be written explicitly (Dutta et al., 2016).

Those exact branches include qualitatively distinct cosmologies. For $\sigma$2 in the Noether-selected polynomial potential, one branch with $\sigma$3 yields a bouncing universe, with contraction for $\sigma$4, a bounce near $\sigma$5, and expansion for $\sigma$6; another branch with $\sigma$7 is described as big-bang-like with later power-law expansion. For $\sigma$8, the scale factor reduces to

$\sigma$9

corresponding to dust-like FLRW behavior (Dutta et al., 2016). These solutions illustrate a recurrent theme in coupled quintom models: the same two-field sector can support de Sitter, matter-like, bouncing, and phantom-crossing branches, depending on the symmetry-selected interaction and the integration constants.

4. Nonminimal, teleparallel, and modified-gravity couplings

A second major branch of the subject generalizes the minimal two-field picture by coupling the quintom sector to geometric invariants. In generalized teleparallel quintom dark energy, each field is nonminimally coupled both to the torsion scalar $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$0 and to the boundary term $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$1, using four coupling functions $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$2, $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$3, $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$4, and $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$5 in an action built around the identity $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$6. With quadratic couplings $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$7, $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$8, $$S=\frac{1}{2}\int d^4x \sqrt{-g}\Big[R+\phi_{;\mu}\phi^{;\mu} -\sigma_{;\mu}\sigma^{;\mu}-2V(\phi,\sigma)\Big],$$9, and $$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$0, and separable exponential potentials, the phase space becomes six-dimensional and contains 21 critical points, 13 of them physical. The analysis identifies several dark-energy-dominated frozen-field solutions and de Sitter-like critical lines $$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$1 and $$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$2 with $$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$3 and $$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$4, while the numerical evolution shows matter domination followed by dark-energy domination and late-time accelerated expansion close to de Sitter; both pure-$$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$5- and pure-$$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$6-coupled submodels cross $$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$7, and the $$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$8-coupled case shows more pronounced oscillations (Bahamonde et al., 2018).

A related curvature-coupled construction couples a quintom pair directly to the Ricci scalar through

$$\mathcal{L}=-3a\dot a^2+a^3\left(\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma)\right).$$9

with exponential-squared potentials

$$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$0

The dynamical-system analysis finds radiation-like, matter-dominated, and de Sitter critical structures, and the model is presented as a viable scalar-tensor background history in which the couplings $$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$1 reshape the phase portrait and affect phantom-divide crossing (Marciu, 2020).

More recent modified-gravity constructions fold quintom matter into higher-curvature or matter-trace dynamics. In an $$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$2-quintom framework, the scalar sector

$$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$3

is combined with a generalized geometric action $$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$4. The paper identifies $$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$5 as phantom and $$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$6 as canonical, rewrites the cosmology in effective-fluid form, and claims a double phantom-divide-line crossing of $$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$7 during a nonsingular bounce, together with $$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$8 in the numerical examples and a degeneracy condition for suppressing higher-derivative pathologies at FLRW level (Milani, 1 Oct 2025).

5. Single-component and nonstandard coupled realizations

The phrase “coupled quintom model” has also acquired a broader meaning in work that realizes quintom behavior without the standard two-scalar potential coupling. One prominent example places a single fluid or single scalar into Nieh–Yan modified teleparallel gravity and couples it directly to the Nieh–Yan density. The central result is that the Nieh–Yan term vanishes on homogeneous FRW, so the background evolution remains that of the uncoupled fluid or scalar, but the perturbation theory changes radically: the antisymmetric teleparallel equation enforces

$$\rho=\frac12\dot\phi^2-\frac12\dot\sigma^2+V(\phi,\sigma),\qquad p=\frac12\dot\phi^2-\frac12\dot\sigma^2-V(\phi,\sigma),$$9

equivalently $$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$0, so the dangerous dark-energy scalar perturbation ceases to propagate. In the examples given, the fluid realization produces a CPL-like history with $$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$1, $$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$2, $$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$3, and crossing near $$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$4, while the scalar realization crosses near $$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$5 with $$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$6 (Kang et al., 31 Jan 2026).

A different nonstandard realization is spinor quintom cosmology with intrinsic spin in Einstein–Cartan–Sciama–Kibble gravity. Here the relevant coupling is geometric: the spinor’s intrinsic spin sources torsion, producing an axial–axial self-interaction in the Dirac equation,

$$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$7

The effective energy density and pressure become

$$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$8

so crossing occurs when

$$w=\frac{p}{\rho} =\frac{\frac12\dot\phi^2-\frac12\dot\sigma^2-V}{\frac12\dot\phi^2-\frac12\dot\sigma^2+V}.$$9

The paper presents three crossing scenarios and argues that intrinsic spin stabilizes the pressure, avoids Big Rip behavior, and can drive the system toward a matter-like epoch with $$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$0 in some branches (Dil, 2016).

An even more unconventional internal dark-sector coupling emerges in a 5D non-perturbative gauge-Higgs unification construction. There the late-time 4D effective theory contains a physical scalar, phantom scalar, physical gauge mode, gauge ghost, and $$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$1-ghost sector generated by dimension-6 operators near a finite localization cutoff. The model has no polynomial scalar potential; the quintom behavior is driven by kinetic, mass-like, and higher-derivative terms, and the massive gauge ghost is identified as crucial for naturally producing Quintom-B histories. In the parameter scan reported, roughly half of the effective parameter space yields Quintom-B evolution with crossing around $$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$2, $$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$3 (Koutroulis, 25 Mar 2026).

6. Phenomenology, observational motivation, and conceptual issues

The phenomenological hallmark of a coupled quintom model is phantom-divide crossing, but the concrete realization varies widely. In the symmetry-selected interacting model

$$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$4

the numerical evolution was reported to show a Hubble history resembling $$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$5CDM, an oscillatory $$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$6 that crosses $$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$7, late-time phantom domination, and stable attractor behavior (Aslam et al., 2013). In teleparallel coupled quintom cosmology, both pure-$$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$8- and pure-$$\ddot\phi+3H\dot\phi+V_{,\phi}=0,\qquad \ddot\sigma+3H\dot\sigma-V_{,\sigma}=0,$$9-coupled submodels pass through matter domination into dark-energy domination and late acceleration, with $\sigma$0 asymptotically and more pronounced oscillations in the boundary-coupled case (Bahamonde et al., 2018).

The observational impetus for quintom model building has sharpened in the DESI era. The staged review of quintom cosmology after DESI DR2 emphasizes that a single perfect fluid or a single minimally coupled scalar with $\sigma$1 cannot smoothly cross $\sigma$2, and it organizes viable models into two-field, higher-derivative, modified-gravity, and EFT classes. The same review also discusses direct derivative couplings of a dynamical quintom field to matter currents,

$\sigma$3

and a Chern–Simons coupling to photons,

$\sigma$4

as possible interaction channels beyond the purely gravitational sector (Cai et al., 30 May 2025).

A concrete late-time application is the Hubble-tension-motivated model with one canonical field $\sigma$5 and one phantom field $\sigma$6, where only the phantom is conformally coupled to matter. In that model,

$\sigma$7

so the phantom sector behaves as a negative-energy stiff component. The benchmark scans reported that the model can outperform $\sigma$8CDM in part of parameter space, specifically $\sigma$9 and $V(\phi,\sigma)$00, while significantly relieving the Hubble tension even though not completely resolving it (Panpanich et al., 2019).

Several conceptual issues remain central. First, the standard no-go theorem keeps the minimal single-fluid or single-$V(\phi,\sigma)$01 scalar model from crossing smoothly, so coupled quintom constructions necessarily rely on extra fields, higher derivatives, or modified geometry (Cai et al., 30 May 2025). Second, the 2015 reassessment of Noether gauge symmetry shows that some earlier symmetry claims were incorrect, so formal integrability results must be checked with care (Ali, 2015). Third, in nonstandard single-component realizations, the coupling often functions by removing the dangerous perturbation mode rather than by making a pathological mode healthy; the Nieh–Yan construction is explicit on this point (Kang et al., 31 Jan 2026). A plausible implication is that “coupled quintom model” is now less a single model class than a family of mechanisms for realizing $V(\phi,\sigma)$02 crossing while controlling the associated dynamical obstruction.