A direct derivation of an effective Hamiltonian in non-relativistic quantum electrodynamics

Abstract: We present a direct derivation of Arai's effective Hamiltonian in non-relativistic quantum electrodynamics without relying on the scaling limit. Our result applies to a broader class of potentials, including the Rollnik class and confining potentials such as the harmonic potential.

• The paper introduces a rigorous derivation of the effective Hamiltonian for a particle in non-relativistic QED, bypassing traditional scaling limit methods.
• It employs dressed electron states and quadratic form techniques to connect microscopic field coupling with observable energy shifts, exemplified by the Lamb shift.
• The study broadens the theory’s applicability to various external potentials, including unbounded and confining forms, ensuring self-adjointness and spectral control.

Direct Derivation of an Effective Hamiltonian in Non-Relativistic QED

Introduction and Motivation

This work develops a rigorous, direct derivation of the effective Hamiltonian for a particle coupled to a quantized radiation field within non-relativistic QED, sidestepping the traditional reliance on scaling limit arguments. The context is the Pauli–Fierz model (with the dipole approximation and omitting the $\mathbf{A}^2$–term) describing a charged particle, such as an electron, subject to an external potential and the electromagnetic field. The effective Hamiltonian captures how vacuum fluctuations, through the mechanism elucidated by Welton for the Lamb shift, average the external potential, thereby affecting observable energy levels. Previously, Arai derived this effective Hamiltonian via a scaling limit approach ("An asymptotic analysis and its application to the nonrelativistic limit of the Pauli–Fierz and a spin-boson model" [MR1075749]), but that methodology requires restrictive assumptions on the potentials and regularity. This paper generalizes the derivation to a much broader class of external potentials, including the Rollnik class and confining (e.g., harmonic) potentials.

Theoretical Framework and Model Specification

The Pauli–Fierz Hamiltonian is considered in $d\ge 2$ dimensions with Hilbert space $$\mathscr{H}_\mathrm{tot} = L^2(\mathbb{R}^d)\otimes\mathscr{F}_b(\mathscr{H}_\mathrm{ph})$$, where the bosonic field is quantized and coupled minimally to the particle’s momentum. The Hamiltonian is

$H_\mathrm{tot} = H_0 \dot{+} (V\otimes 1),$

where $H_0$ involves the kinetic term (with renormalized mass), field energy, and interaction $-\frac{q}{m}\sum p_j \otimes A_j(0)$. The class of potentials $V$ includes, besides traditional Rollnik-class potentials, unbounded and confining forms such as the harmonic potential, covered by precise form-boundedness and integrability assumptions. Important physical input is the renormalization linking bare and observed mass parameters, ensuring that the ground state energy of each fiber in the field’s momentum decomposition is correctly normalized.

Dressed Electron States and Effective Hamiltonian Construction

The essential step is the construction of dressed electron states, representing superpositions of fiber Hamiltonian ground states parametrized by electron momentum, such that

$$\psi_\mathrm{dr}(\mathbf{x}) = \frac{1}{(2\pi\hbar)^{d/2}} \int (\mathcal{F}_d\psi)(\mathbf{P})\, \Phi_0(\mathbf{P}) e^{i\mathbf{x}\cdot\mathbf{P}/\hbar} \,d\mathbf{P}$$

for a wavefunction $\psi \in L^2(\mathbb{R}^d)$. The fiber decomposition and the explicit diagonalization of $H_0$ via a van Hove transformation ensure analytic control over the structure of the field-coupled states. The dressed states are the canonical states for calculating physical expectation values, as they encode the interaction with the vacuum.

Through a detailed analysis of quadratic forms, the paper precisely establishes the mapping between the domain structures of the Pauli–Fierz Hamiltonian and the emergent effective operator in $d\ge 2$0. The effective Hamiltonian is defined as the unique self-adjoint operator whose quadratic form expectation matches that of $d\ge 2$1 computed in the dressed state: $d\ge 2$2 where the effective potential is

$d\ge 2$3

and $d\ge 2$4 depends explicitly on model parameters and the cutoff function $d\ge 2$5, incorporating the field-coupling structure.

Main Results and Domain Analysis

The work rigorously justifies the existence and self-adjointness of both $d\ge 2$6 and $d\ge 2$7 for the extended class of potentials. Notably, it is shown that for potentials bounded from below or infinitesimally form-bounded with respect to $d\ge 2$8, the quadratic form sum method applies, ensuring that both operators are well-defined and bounded from below.

The characterization via quadratic form expectations ensures that spectral information about $d\ge 2$9 directly bounds and relates to the spectrum of $$\mathscr{H}_\mathrm{tot} = L^2(\mathbb{R}^d)\otimes\mathscr{F}_b(\mathscr{H}_\mathrm{ph})$$0: $$\mathscr{H}_\mathrm{tot} = L^2(\mathbb{R}^d)\otimes\mathscr{F}_b(\mathscr{H}_\mathrm{ph})$$1 providing rigorous control over the spectral shift induced by field coupling.

A concrete example is furnished by the harmonic oscillator: when $$\mathscr{H}_\mathrm{tot} = L^2(\mathbb{R}^d)\otimes\mathscr{F}_b(\mathscr{H}_\mathrm{ph})$$2, the effective potential is $$\mathscr{H}_\mathrm{tot} = L^2(\mathbb{R}^d)\otimes\mathscr{F}_b(\mathscr{H}_\mathrm{ph})$$3, thus replicating the physically anticipated uniform energy shift, manifesting the Lamb shift structure.

Mass Renormalization and Operator Structure

A notable point is the explicit treatment of mass renormalization in the kinetic and interaction terms. The analysis distinguishes between the physical scenarios where either the bare or renormalized mass appears in the interaction term, affecting the effective potential’s structure. The residual dependence of the effective potential on the bare mass, in certain model variations, is highlighted and attributed to the omission of the $$\mathscr{H}_\mathrm{tot} = L^2(\mathbb{R}^d)\otimes\mathscr{F}_b(\mathscr{H}_\mathrm{ph})$$4 term, a non-trivial aspect in infrared QED modeling. The direct method also elucidates how the effective Hamiltonian’s structure reflects the modeling approximation, an insight critical for foundational studies and practical modeling applications.

Implications and Directions for Future Work

This direct quadratic form approach provides a robust framework for deriving and analyzing effective Hamiltonians in QED settings with singular, unbounded, or confining external potentials. The generalization beyond the scaling limit significantly increases the modeling flexibility for atomic, molecular, and condensed matter systems where vacuum-induced potential smoothing is relevant. Additionally, the methodology is positioned to adapt to further generalizations, such as situations with non-negligible $$\mathscr{H}_\mathrm{tot} = L^2(\mathbb{R}^d)\otimes\mathscr{F}_b(\mathscr{H}_\mathrm{ph})$$5 terms (cf. Hiroshima [MR1235953, MR1438035]) and many-particle or higher-order interaction scenarios.

The theoretical advances here interface with problems in the rigorous analysis of the Lamb shift and spectral theory of QED models [MR2770092], opening the route for future analyses of binding energies, localization, or quantum control in strongly coupled light-matter systems.

Conclusion

The paper provides a mathematically rigorous, direct derivation of the effective Hamiltonian for the non-relativistic Pauli–Fierz model, applicable to a substantially larger class of external potentials than available in previous work. By characterizing the effective operator via dressed state quadratic forms, the approach clarifies the link between microscopic field coupling and effective particle dynamics, with broad implications for spectral analysis and potential extensions to models incorporating more general quantum field interactions (2604.22316).