Frequency-Dependent Fluctuating Charges (ωFQ)

• ωFQ is a model that defines atomic charges dynamically by coupling them to external fields at frequency ω, extending traditional static schemes.
• It integrates classical conduction and quantum tunneling effects through coupled linear systems to capture non-static polarization and transport phenomena.
• The framework underpins efficient simulations in nanoplasmonics, electrochemistry, and quantum transport, enabling predictions of optical responses and impedance behaviors.

Frequency-dependent fluctuating charges ($\omega$FQ) refer to atomic or molecular charge degrees of freedom whose dynamics are explicitly coupled to an external field at frequency $\omega$, yielding a complex, non-static polarization response. These models extend traditional fluctuating charge schemes into the frequency domain, enabling atomistic or coarse-grained descriptions of optical, charge–transfer, and transport phenomena that depend on both geometry and the frequency of the applied perturbation. The $\omega$FQ paradigm underpins several modern approaches in nanoplasmonics, quantum transport, electrochemistry, and polarizable force fields.

1. Mathematical Formulation of Frequency-Dependent Fluctuating Charges

The $\omega$FQ framework is based on assigning to each atom (or site) a complex frequency-dependent charge $q_i(\omega)$, governed by a coupled linear system that incorporates both classical conduction and quantum tunneling effects. In atomistic models for plasmonic nanostructures, such as those in plasmonX (Giovannini et al., 14 Oct 2025), the canonical system is

$$\big[\,\overline{\mathbf{K}}\;\mathbf{T}^{qq} - z_q(\omega)\,\mathbf{I}_N\,\big]\,\mathbf{q}(\omega) = -\overline{\mathbf{K}}\,\mathbf{V}(\omega)$$

where $\overline{\mathbf{K}}$ encodes interatomic coupling and quantum tunneling (via a Fermi-like damping function), $\mathbf{T}^{qq}$ is the charge–charge interaction kernel (often based on Gaussian integrals to account for charge delocalization), $z_q(\omega)$ captures Drude damping, and $\mathbf{V}(\omega)$ represents the external field.

The key innovation lies in the frequency dependence: charge redistribution is no longer instantaneous but governed by the continuity equation and a relaxation time (or scattering time) $\tau_i$ per atom,

$$-i\omega q_i(\omega) = \sum_j \frac{A_j n_j (1 - f_{ji}(l_{ij}))}{1/\tau_j - i\omega} \frac{\phi_i(\omega) - \phi_j(\omega)}{l_{ij}} + \dots$$

with $f_{ji}(l_{ij})$ suppressing charge transfer across large atomic distances (modeling tunneling).

Interband polarization response is captured by adding frequency-dependent dipole variables $\mu_i(\omega)$ governed by atomic polarizabilities $\alpha_i^{\text{IB}}(\omega)$, such that

$$\mu_i(\omega) = \alpha_i^{\mathrm{IB}}(\omega) [E_i^{q}(\omega) + E_i^{\mu}(\omega) + E_i^{\mathrm{ext}}(\omega)]$$

These coupled systems yield atom-resolved charge/dipole distributions and allow direct calculation of frequency-dependent polarizability, absorption cross sections, and field-induced density patterns.

2. Physical Interpretation and Theoretical Characterization

Frequency-dependent fluctuating charge models represent an advancement beyond static polarizable models such as QEq and QTPIE (Chen, 2010), which solve for $q_i$ via constrained quadratic minimization: $$E(q) = \sum_{i} X_i q_i + \frac{1}{2} \sum_{ij} J_{ij} q_i q_j$$ subject to $\sum_i q_i = Q$ (charge conservation), with $J_{ij}$ incorporating atomic hardness and screened Coulomb interactions.

While these models efficiently capture polarization and charge transfer, they lack dynamic response; i.e., their charge distributions are strictly static or adiabatic. The $\omega$FQ extension introduces explicit time/frequency dependence, which is crucial for non-equilibrium phenomena (e.g., AC transport in quantum dots (Moca et al., 2010), frequency-dependent Casimir energies (Graham et al., 2014), charge fluctuation forces in nanocircuits (Drosdoff et al., 2015), and electrochemical impedance (2206.13322)).

Importantly, $\omega$FQ models preserve spatial symmetries such as translational invariance and correctly recover physical properties like dipole moments, polarizability tensors, and their frequency-dependent counterparts when proper charge-conserving terms are included.

3. Applications in Nanoplasmonics, Molecular Response, and Electrochemistry

Atomistic $\omega$FQ and $\omega$FQF$\mu$ models have demonstrated utility in simulating the optical response of metal and graphene-based nanostructures (Giovannini et al., 14 Oct 2025, Nicoli et al., 2024). They accurately reproduce plasmon resonance frequency shifts, intensity modulations in alloys (linear Vegard’s law behavior), and field-induced charge patterns in various geometries—approaching ab initio accuracy with orders-of-magnitude lower computational cost.

In multiscale frameworks, such as the $\omega$FQF$\mu$/FQ model (Nicoli et al., 2024), the plasmonic nanoparticle is treated with frequency-dependent charges and dipoles, and the explicit solvent is modeled with FQ force fields. Mutual polarization is accounted for via dynamical couplings, enabling the calculation of solvatochromic red shifts and the influence of solvent composition (e.g., water/ethanol mixtures) on optical properties.

Molecular response to dynamic fields is rigorously handled in models such as ACKS2$\omega$ (Cheng et al., 2022), where the frequency-dependent molecular polarizability and $C_6$ dispersion coefficients are computed via partitioning the response function, retaining full frequency dependence in the non-interacting response kernel and reusing an adiabatic (frequency-independent) hardness matrix.

In electrochemistry, constant-potential molecular dynamics (2206.13322) link the frequency-dependent impedance $Z(\omega)$ and capacitance to time-correlations of electrode charge fluctuations: $$Y(\omega) = \beta \Big[ i\omega \langle\delta Q^2\rangle + \omega^2 \int_0^\infty \langle\delta Q(0)\,\delta Q(t)\rangle e^{-i\omega t} dt \Big]$$ providing direct microscopic interpretation of impedance spectroscopy in nanocapacitors and offering a bridge between equilibrium fluctuations and linear response.

4. Role in Quantum Noise and Conductivity

In quantum transport, models of $\omega$FQ manifest as frequency-dependent noise spectra in systems such as quantum dots (Moca et al., 2010) and ensembles of two-level fluctuators (You et al., 2020). Real-time functional renormalization group (FRG) analysis reveals non-local (retarded) current vertices, resulting in sharp anti-resonances in the noise spectrum at $\hbar \omega = \pm eV$, directly linked to the dynamical fluctuation of charges.

In ionics and electrolytes, stochastic density functional theory (SDFT) (Bonneau et al., 2024) generalizes the classical Debye–Falkenhagen result for AC conductivity to concentrated solutions, by linearizing around fluctuating charge densities and including both Coulombic and hard-core repulsion. The non-equilibrium conductivity is determined by the interplay between field frequency and the relaxation time of the ionic cloud, leading to frequency-dependent corrections: $$\kappa(\omega) = \kappa_0 + \kappa_\text{el}(\omega) + \kappa_\text{hyd}(\omega)$$ with $\kappa_\text{el}(\omega)$ reflecting the incomplete distortion of the ionic cloud under high-frequency fields—a direct dynamical manifestation of $\omega$FQ principles.

5. Parameter Dependence, Numerical Methods, and Implementation

The strength and behavior of frequency-dependent fluctuating charges are critically dependent on microscopic parameters such as:

• Atomic area $A_i$, electron density $n_i$, and relaxation time $\tau_i$ (reflecting plasmonic material properties)
• Quantum capacitance $C_Q$ and geometrical capacitance $C_0$, especially in graphene-based nanostructures (Drosdoff et al., 2015)
• Response kernels $T^{qq}$ and $T^{q\mu}$ (charge–dipole coupling)
• Fermi-like damping parameters, influencing tunneling rates
• System geometry, solvent composition, and temperature (modulating charge fluctuation amplitude and the associated interactions)

Numerically, $\omega$FQ models are formulated as block linear systems of dimension $N$ (or $4N$ for charge plus dipole models) and benefit from explicit mappings between bond and atom space (Chen, 2010), ensuring computational efficiency and correct symmetry properties. Singular value decomposition (SVD), complete orthogonal decomposition (COD), and graph-theoretic incidence matrix analysis are applied to address rank deficiency and constraint enforcement.

Implementation is carried out in open-source packages such as plasmonX (Giovannini et al., 14 Oct 2025), which supports both atomistic and boundary element methods, and includes post-processing modules for visualization of induced densities and fields.

6. Limitations and Conceptual Challenges

While $\omega$FQ models successfully capture dynamic polarization and charge transfer, several open challenges remain:

• Overestimation of long-range charge transfer in traditional fluctuating charge models and its correction via distance-dependent electronegativities (as in QTPIE)
• Ambiguities in field coupling prescriptions (E′ vs. E″ routes), particularly in nonlocal models
• Ensuring size extensivity and correct asymptotic scaling of polarizability and other observables (noted in the failure of reparameterized QEq vs. handling in QTPIE)
• Treatment of dissipation and statistical fluctuations, especially in Casimir energy problems involving frequency-dependent dielectric functions (Graham et al., 2014); necessitating careful choice between field-theoretical and statistical mechanical frameworks to correctly account for extra terms and cancellations

Some models remain adiabatic (hardness kernel frequency-independent), and thus may not capture nonlocal frequency-dependent correlation effects (for example, dynamic damping in $f_{xc}(n,q,\omega)$ (Ruzsinszky et al., 2020)). There is also a challenge in bridging classical fluctuating charge models with fully quantum mechanical time-dependent density functional theory (TDDFT) approaches.

7. Outlook and Applications

The $\omega$FQ paradigm underpins atomistic simulation of dynamic optical and transport properties in metal/graphene nanostructures, explicit solvents, and electrolytes; it enables computational modeling at scales and costs inaccessible to ab initio quantum methods, while retaining key quantum and statistical physical effects.

Continued development focuses on:

• Integration with polarizable force fields (ACKS2$\omega$ (Cheng et al., 2022))
• Coupling with multiscale models for solvated nanoplasmonic systems (Nicoli et al., 2024)
• Extension to include dynamic field coupling, dissipation, and retardation effects in complex environments
• Deployment in open-source atomistic simulation codes (e.g., plasmonX (Giovannini et al., 14 Oct 2025)) with comprehensive post-processing for charge density and field analysis

The frequency-dependent fluctuating charge model provides an essential, physically grounded, and computationally efficient platform for nanoplasmonics, molecular polarization, electrochemical dynamics, and quantum transport applications. Its continued refinement is likely to yield ever more accurate, extensible, and scalable physical models for emergent dynamical phenomena across disciplines.