Dynamical Polaron-Field Model

• Dynamical polaron-field models are coupled matter–field systems where particles such as electrons create a self-induced medium response that renormalizes their dispersion and effective mass.
• They employ diverse methodologies, including strong-coupling reductions, variational closures like Dirac–Frenkel, and asymptotic analyses that reveal retarded self-interactions with temporal memory.
• These models span continuum, lattice, and ultracold gas settings, providing practical insights into phonon-mediated transport, nonequilibrium dynamics, and field-induced feedback mechanisms.

A dynamical polaron-field model is a coupled matter–field description in which an electron, exciton, or impurity evolves together with a medium degree of freedom generated by its own motion, such as a quantized phonon field, a coherent phonon background, a classical polarization field, a condensate order parameter, or a local dynamical self-energy. Across the Fröhlich, Holstein, SSH, mean-field, and Bose-polaron settings, the common structure is self-consistent backaction: the particle induces a deformation or excitation field, and that field in turn renormalizes dispersion, mass, transport, spectra, and relaxation. In the cited literature, this class of models includes strong-coupling Fröhlich dynamics, Landau–Pekar and Schrödinger–Poisson reductions, driven Holstein polarons, inhomogeneous DMFT formulations, and impurity dynamics in Bose condensates (Frank et al., 2013, Frank et al., 2015, Griesemer, 2016, Griesemer et al., 2016, Richler et al., 2018).

1. Microscopic formulations

In the continuum Fröhlich setting, the dynamical model describes one electron interacting with a quantized phonon field in a polar crystal. In strong-coupling units, the Hamiltonian is

$$H_\alpha = p^2 +\int_{\mathbb R^3}\frac{dk}{|k|} \left(e^{-ikx}a_k+e^{ikx}a_k^*\right) +\int_{\mathbb R^3}dk\, a_k^*a_k,$$

with $p=-i\nabla_x$ and $\alpha$-dependent canonical commutation relations

$[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$

The scaling $\alpha\to\infty$ suppresses quantum fluctuations of the field and is the basis of the strong-coupling semiclassical picture (Frank et al., 2013).

A distinct continuum realization is the mean-field polaron model,

$$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$

where $u$ is the electron wave function and $v$ is the self-generated electrostatic potential. Here the field equation is a second-order oscillator equation with a small prefactor $\varepsilon^2$, so the fast-field limit $\varepsilon\to 0$ is singular rather than perturbative (Griesemer et al., 2016).

On lattices, the dynamical polaron-field model appears in several forms. The inhomogeneous Holstein Hamiltonian,

$p=-i\nabla_x$0

describes a single electron coupled locally to dispersionless phonons in a spatially inhomogeneous environment. A related mixed-coupling lattice model combines diagonal Holstein coupling and non-diagonal SSH coupling through a general electron-boson Hamiltonian with interaction vertex $p=-i\nabla_x$1, where $p=-i\nabla_x$2 depends on electron momentum because it modulates hopping (Richler et al., 2018, Marchand et al., 2016).

In ultracold-gas realizations, the field need not be a crystal phonon field. For a single impurity in a weakly interacting one-dimensional Bose gas, the microscopic Hamiltonian is

$p=-i\nabla_x$3

and after the Lee–Low–Pines transformation the condensate profile is described in the impurity frame (Will et al., 2023). In a dense Bose gas with one impurity, the many-body Hamiltonian

$p=-i\nabla_x$4

yields, after condensate/excitation separation and Bogoliubov reduction, an effective impurity–phonon model of Bogoliubov–Fröhlich type (Lampart et al., 13 Apr 2026).

2. State manifolds, coherent fields, and variational closures

A central structural choice in dynamical polaron-field models is the manifold on which the coupled evolution is projected or approximated. In the strong-coupling Fröhlich problem, Pekar’s ansatz takes the initial state as a product of an electron wave function and a phonon coherent state,

$p=-i\nabla_x$5

with Weyl operator $p=-i\nabla_x$6. The coherent shift identity

$p=-i\nabla_x$7

makes $p=-i\nabla_x$8 a macroscopic coherent background in strong-coupling units (Frank et al., 2013).

The Dirac–Frenkel time-dependent variational principle gives a geometric formulation of this product-state reduction. On the manifold

$p=-i\nabla_x$9

the projected evolution satisfies $\alpha$0, and for Hamiltonians of the form $\alpha$1 this produces coupled equations for the particle factor and the field factor. In the Fröhlich case, the electron equation is Schrödinger-like, while the phonon equation is linear and can be solved explicitly; eliminating the field generates a retarded self-interaction with memory (Griesemer, 2016).

Lattice formulations often use coherent-state trial manifolds of Davydov type. For the one-dimensional Holstein polaron, the Davydov $\alpha$2, $\alpha$3, and $\alpha$4 ansatzes approximate the coupled exciton–phonon dynamics through site amplitudes and coherent phonon displacements, while the multi-$\alpha$5 ansatz uses a superposition of $\alpha$6 $\alpha$7 components. In the electric-field-driven ring problem, the multi-$\alpha$8 ansatz is propagated by the Dirac–Frenkel principle and benchmarked against the hierarchy equations of motion; the reported relative deviation obeys

$\alpha$9

so the variational dynamics becomes numerically exact in the large-$[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$0 limit (Sun et al., 2010, Huang et al., 2017).

Other closures encode the field indirectly. In the two-band impurity-in-BEC problem, an extended two-band Lang–Firsov transformation introduces variational dressing matrices $[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$1 and produces a coherent polaron Hamiltonian with renormalized hopping, shifted on-site energies, and induced long-range interactions (Yin et al., 2015). In inhomogeneous DMFT, the closure is a local, site-dependent self-energy,

$[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$2

which retains full frequency dependence while suppressing off-diagonal self-energy structure (Richler et al., 2018).

3. Effective equations in asymptotic regimes

The strongest rigorous results concern asymptotic limits in which the field becomes effectively classical, frozen, or constrained. For the dynamical Fröhlich polaron in the strong-coupling limit, if the initial state is exactly $[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$3, then

$[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$4

where

$[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$5

To leading order on the natural time scale, the phonon coherent state is stationary and the electron evolves by the effective linear Schrödinger equation $[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$6. The bound remains meaningful up to $[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$7 (Frank et al., 2013).

A more refined strong-coupling result derives the Landau–Pekar system directly from the Fröhlich Hamiltonian. The effective equations are

$[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$8

with $[a_k,a_{k'}^*]=\alpha^{-2}\delta(k-k').$9 generated by the classical field. For initial data $\alpha\to\infty$0, the exact state is approximated in norm by

$\alpha\to\infty$1

with remainder bounded by

$\alpha\to\infty$2

while the particle and field reduced density matrices are approximated with $\alpha\to\infty$3 trace-norm error on the same time interval (Frank et al., 2015).

Within the Dirac–Frenkel framework, eliminating the phonon factor yields a nonlinear, nonlocal Schrödinger equation with memory. The effective potential satisfies

$\alpha\to\infty$4

and stationary reduction recovers the Pekar equation. For coherent initial data associated with a Pekar minimizer, the full quantum dynamics is approximated with

$\alpha\to\infty$5

and the method extends to $\alpha\to\infty$6-polaron systems (Griesemer, 2016).

In the weak-coupling or fast-field regime, the effective closure is different. Setting $\alpha\to\infty$7 in the mean-field model gives $\alpha\to\infty$8, hence the Schrödinger–Poisson system or equivalently the Choquard equation

$\alpha\to\infty$9

If the initial data satisfy the $$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$0 preparation condition, then on a fixed interval $$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$1

$$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$2

The proof relies on oscillatory cancellation and integration by parts in time to remove the dangerous $$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$3 factor (Griesemer et al., 2016).

Additional asymptotic reductions show how singular or collective fields can emerge. In a quasi-classical three-dimensional polaron coupled to acoustic and optic phonons, specially prepared coherent states produce an effective time-dependent point interaction with coupling

$$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$4

and the microscopic dynamics converges in strong operator topology to the unitary evolution generated by the time-dependent point interaction (Carlone et al., 2019). In the dense Bose-gas setting, the microscopic impurity problem converges in the joint large-density/large-volume limit to the translation-invariant Bogoliubov–Fröhlich Hamiltonian

$$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$5

with Bogoliubov dispersion

$$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$6

and an explicit $$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$7-convergence rate under $$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$8, $$i\partial_t u = -\Delta u + vu,\qquad \varepsilon^2 \partial_t^2 v = -v + \Delta^{-1} |u|^2,$$9 (Lampart et al., 13 Apr 2026).

4. Response theory and driven nonequilibrium dynamics

Dynamical polaron-field models are particularly informative out of equilibrium because they distinguish bare driving from field-mediated feedback. In the truncated phase space approach to linear response, the external electric field induces a retarded internal field through the self-energy. The current obeys

$u$0

with $u$1. Because the extra term enters as $u$2, the dc mobility is unchanged to first order, whereas the effective mass and optical absorption are modified already at first order. For the Fröhlich polaron at $u$3,

$u$4

so the dynamical screening reduces the first-order mass enhancement by a factor of two relative to the no-screening case (Sels et al., 2014).

For a one-dimensional Holstein polaron in a constant electric field switched on at $u$5, the nonequilibrium dynamics exhibits damped Bloch oscillations, field-induced phonon emission, and in suitable regimes a steady current. The exact energy-gain sum rule,

$u$6

shows that the electric work is converted into phonon production. In weak coupling, the steady-state current at large field is reported to scale as

$u$7

with $u$8, a behavior interpreted in terms of coherent propagation between Stark states rather than incoherent hopping. In steady state, the moving polaron leaves behind a phonon trail whose average density obeys $u$9 (Vidmar et al., 2010).

On a finite ring, the field-driven Holstein polaron behaves differently. Starting from a broad wave packet, the exciton undergoes Bloch oscillations, and weak exciton–phonon coupling broadens the packet and reduces the current amplitude while leaving the temporal periodicity unchanged. Starting from a narrow packet, the uncoupled exciton shows a symmetric breathing mode, but weak coupling breaks that symmetry and produces a non-zero exciton current. The paper emphasizes that, at variance with the case of an infinite linear chain, no steady state is found in a finite-sized ring within the anti-adiabatic regime (Huang et al., 2017).

In the Holstein–Hubbard Mott insulator driven by strong DC or pulsed electric fields, the field does not merely heat the system. For $v$0 in the paper’s units, strong DC drive induces a quasi-steady metallic state in which field-induced doublon–hole production is balanced by phonon-enhanced recombination. Spectrally, the system develops in-gap states identified as dressed doublons and dressed holes, together with Wannier–Stark and phonon sidebands, and the lesser spectrum shows a partial population inversion in the gap region. The paper interprets this as a field-induced localization mechanism that effectively enhances the phonon coupling; the relaxation scale is found to be approximately controlled by

$v$1

Time-resolved photoemission and time-resolved optical conductivity are proposed as probes of these nonequilibrium polaronic features (Werner et al., 2014).

5. Lattice, inhomogeneous, and multiband generalizations

One important branch of the subject replaces the usual mass-renormalization narrative by a band-reshaping narrative. In the mixed Holstein/SSH model, the low-energy polaron can be described by an effective band

$v$2

where SSH-type coupling generates longer-range hoppings $v$3. In the pure SSH limit, the ground state changes from $v$4 to finite momentum at a critical coupling $v$5, with $v$6 in the anti-adiabatic limit. With both Holstein and SSH couplings, this becomes a transition line in coupling space, and increasing $v$7 generally lowers $v$8. The paper stresses that this is not a self-trapping transition and not a quantum phase transition: the bandwidth remains finite, the polaron stays mobile, and the singularity is the vanishing of the local curvature at $v$9 (Marchand et al., 2016).

In real-space inhomogeneous systems, the field can be encoded as a local but site-dependent dynamical self-energy. I-DMFT assumes

$\varepsilon^2$0

and represents both $\varepsilon^2$1 and the hybridization $\varepsilon^2$2 as continued fractions obtained by Haydock recursion or Lanczos tridiagonalization. The method is exact in the non-interacting limit $\varepsilon^2$3 and reduces to standard DMFT in the homogeneous Holstein limit. For a defect potential $\varepsilon^2$4, the computed local density of states shows Friedel oscillations,

$\varepsilon^2$5

whose wavelength is controlled by the polaron effective mass through $\varepsilon^2$6. The framework is proposed as a direct interpretation tool for STM in systems with sizable electron–lattice interactions (Richler et al., 2018).

The standing-polaron problem on a one-dimensional harmonic lattice gives a mixed quantum–classical realization of the dynamical model. The lattice displacements are classical, the electron is quantum, and the SSH coupling modulates the hopping through bond deformation. In dimensionless form,

$\varepsilon^2$7

Minimization yields the self-consistent distortion $\varepsilon^2$8, and the model supports both small-radius and large-radius standing polarons. In the broad-polaron regime, the continuum profile is

$\varepsilon^2$9

The time-dependent formation problem depends strongly on the initial electronic state: localized or uniform initial states do not produce a stable polaron, whereas initial states chosen as low-lying eigenfunctions of the uncoupled lattice do (Likhachev et al., 2012).

Multiband and strong-backaction impurity problems further enlarge the scope of dynamical polaron-field models. In a two-band impurity system immersed in a BEC, the extended Lang–Firsov transformation yields a coherent polaron Hamiltonian with reduced hopping $\varepsilon\to 0$0, shifted band energies $\varepsilon\to 0$1, and phonon-mediated long-range interactions. Residual incoherent processes are treated by a Lindblad master equation with polaron jump operators, and the inter-band relaxation rate is renormalized relative to Fermi’s Golden Rule. In strong coupling, the polaron becomes tightly dressed in each band and cannot tunnel between them, leading to an inter-band self-trapping effect (Yin et al., 2015).

For a mobile impurity in a weakly interacting one-dimensional Bose gas beyond the Fröhlich regime, the field is the condensate itself. The mean-field equation in the co-moving frame,

$\varepsilon\to 0$2

incorporates nonlinear condensate deformation and momentum transfer between impurity and Bose field. The stationary polaron energy becomes periodic in momentum, $\varepsilon\to 0$3, the effective mass $\varepsilon\to 0$4 can diverge at $\varepsilon\to 0$5, and at larger momentum the impurity velocity can become negative. Dynamically, sudden quenches can generate density waves, grey solitons, branch switching, and even backscattering, while trapped-gas Truncated Wigner simulations identify the parameter regime in which quantum fluctuations remain small (Will et al., 2023).

6. Conceptual issues, misconceptions, and recurring physical themes

A defining feature of the literature is that the “field” in a dynamical polaron-field model does not have a single universal status. In some strong-coupling Fröhlich results, the phonon coherent background is stationary to leading order and the electron sees a frozen potential (Frank et al., 2013). In the Landau–Pekar and Dirac–Frenkel reductions, the field is classical but dynamical, and after elimination it generates a nonlinear Schrödinger equation with temporal memory (Frank et al., 2015, Griesemer, 2016). In the weak-coupling mean-field limit, the field instead tracks the instantaneous density through the algebraic constraint $\varepsilon\to 0$6 and the effective equation is Choquard rather than retarded (Griesemer et al., 2016). In the strong-coupling one-dimensional Bose gas, by contrast, the field is neither frozen nor slaved instantaneously: condensate depletion, soliton emission, and phase winding are essential parts of the dynamics (Will et al., 2023).

Several recurrent misconceptions are explicitly addressed in the cited work. The mixed Holstein/SSH study states that no self-trapping transition exists in the pure Holstein model or in the mixed H/SSH model in a clean system; the transition line marks a shift of the ground-state momentum away from $\varepsilon\to 0$7, not localization or bandwidth collapse (Marchand et al., 2016). The driven Holstein–Hubbard analysis states that the field-induced metallic state with phonons is not the usual infinite-temperature heating of an isolated driven Mott insulator, because phonons provide dissipation and stabilize a quasi-steady current-carrying state (Werner et al., 2014). The finite-ring variational study states that one should not infer the infinite-chain steady state from a finite periodic geometry, since no steady state is found there in the anti-adiabatic regime (Huang et al., 2017).

The observable content of these models is equally varied. In linear response, dynamical screening modifies the effective mass and optical absorption at first order while leaving the dc mobility unchanged to first order (Sels et al., 2014). In real-space DMFT, the impurity-induced LDOS oscillation period carries information about the polaron mass and is directly relevant to STM (Richler et al., 2018). In nonequilibrium Mott systems, time-resolved photoemission and optical conductivity are natural probes of field-induced in-gap polaronic states and phonon sidebands (Werner et al., 2014).

Taken together, the literature shows that the dynamical polaron-field model is not one equation but a family of controlled coupled particle–field descriptions. Its precise form depends on the microscopic medium, the relevant scaling limit, the choice of state manifold or self-energy closure, and the temporal regime of interest. What remains invariant is the central mechanism: a mobile quantum object dresses itself by creating a medium response, and the resulting field feeds back on the object’s subsequent dynamics.