Infrared Problem in Quantum Electrodynamics

Albert Schwarz; Igor Frolov

REVIEW 4 major objections 5 minor 10 references

Infrared Problem in Quantum Electrodynamics

T0 review · 4 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that splitting QED into an exactly solvable eikonal Hamiltonian plus a perturbation makes inclusive cross-section diagrams infrared-finite one by one.

desk verdict A plausible program — 1/m perturbation theory around an exactly solvable eikonal Hamiltonian in the L-functional formalism — but H_as drops the p·k/p⁰ phase that generates the 1/(p·k) poles, so the central IR-finiteness claim fails as written. read the letter →

arxiv 2608.00204 v1 pith:5TAFEWJA submitted 2026-07-31 hep-th math-phmath.MP

Igor Frolov , Albert Schwarz This is my paper · ORCID

classification hep-thmath-phmath.MP

keywords infrareddivergencesinclusivecrosssectionsL-functionalseikonalHamiltoniansoftphotonsquantumelectrodynamics1/mexpansionGGreenfunctions

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to show that the infrared problem of quantum electrodynamics disappears when the theory is reformulated in terms of L-functionals, state functionals on the algebra of creation and annihilation operator exponentials that exist for every representation of the canonical commutation relations. The proposed split writes the QED Hamiltonian as a solvable eikonal Hamiltonian plus a remainder; the eikonal part is the soft-photon (heavy-mass) limit of the interaction and is solved exactly, while the remainder is smaller by one power of photon momentum at low energies. The paper argues that all infrared-dangerous structure sits in the solvable part, so diagrams for inclusive cross sections and inclusive scattering matrices are finite without relying on cancellations. If true, this turns a diagram-by-diagram cancellation story into a diagram-by-diagram finite story, and gives a concrete meaning to the dressed electron's coherent soft-photon cloud outside the usual particle-number Hilbert space.

What carries the argument

The central object is the solvable eikonal Hamiltonian, defined by replacing the QED interaction with a term proportional to (a^+_μ(k)+a_μ(k)) p^μ/p0 ρ(p) plus an instantaneous potential, where ρ(p) is the charge density. In the L-functional formalism the Hamiltonian is doubled (two commuting copies of the field algebra), and it is diagonalized by shifting photon operators by a charge-density-dependent c-number and dressing fermion operators with coherent photon-cloud factors. This exact solution absorbs the whole infrared-singular sector, so perturbation theory in the remainder has vertices vanishing linearly in photon momentum, one power faster than the photon measure, making the diagrams

What would settle it

Compute the soft-photon distribution predicted by the solvable Hamiltonian and compare it with the known classical radiation field of an accelerated charge; if the leading low-momentum behaviour disagrees, the solvable part has not captured the infrared-singular sector and the power counting of Section 5 would not concern the actual IR singularities of QED.

Watch

Extended reading notes

Core claim

The paper claims that in QED formulated with L-functionals, the inclusive scattering matrix is infrared finite order by order in a new perturbation theory. The full Hamiltonian is written as H = H_as + δV, where H_as is the solvable eikonal Hamiltonian whose interaction is built from the charge density times the velocity p^μ/p0 and a non-local instantaneous potential; H_as is the eikonal limit of the true QED interaction and contains the entirety of the infrared-singular structure. The solvable model is solved exactly: shifted photon operators evolve as free photons, and dressed fermion operators propagate freely up to mass renormalization and an infrared-safe soft-photon dressing factor. Th

Load-bearing premise

The load-bearing premise is that the solvable Hamiltonian with a time-independent charge density really is the soft limit of the QED interaction and captures all infrared-divergent structure, even though the true current carries a recoil phase that produces the 1/(p·k) denominators.

Editorial extensions

If this is right

Inclusive cross sections and inclusive scattering matrices in QED are infrared-finite diagram by diagram, with no need to cancel divergences between real and virtual soft-photon contributions.
The physical electron is treated as a dressed state containing a coherent soft-photon cloud; this state is well-defined in the L-functional formalism even though it lies outside the usual particle-number Hilbert space.
Fermionic Green functions of the solvable Hamiltonian equal free ones up to mass renormalization and an infrared-safe dressing factor, while photonic Green functions are exactly the free ones.
The remaining perturbation theory is organized as a 1/m expansion, so every extra order adds more powers of soft momentum and cannot reintroduce a k→0 singularity.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the eikonal split is right, the same strategy may apply to other theories with soft-particle clouds, such as non-abelian gauge theories, though the eikonal Hamiltonian is no longer exactly solvable there and the power counting would need modification.
The paper leaves a direct check unstated: compare the soft-photon distribution produced by the solvable Hamiltonian with the classical radiation field of an accelerated charge; matching the leading behaviour would confirm that the dropped recoil phase is truly subleading.
A practical test of the scheme is to compute the next-to-eikonal correction to a simple bremsstrahlung cross section and compare it with the standard soft-photon expansion; agreement would show the missing recoil phase enters only at the expected order.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Two things to know. The paper runs a real program: split QED into an exactly solvable eikonal Hamiltonian plus a 1/m perturbation, inside the authors' L-functional formalism, and claim inclusive cross sections become IR-finite diagram by diagram. But the solvable Hamiltonian is not the eikonal sector of QED, and the paper half-admits it. Equation (4.5) has the photon phase e^{±iωt} but no electron phase e^{ip·k/p⁰ t}; the sentence after (4.3) concedes that the Kulish–Faddeev asymptotic Hamiltonian 'contains an additional factor depending on time.' That factor is the load-bearing one. Section 3's own calculation shows that integrating the current with phase ω_p − ω produces the 1/(p·k) denominators: the Liénard–Wiechert field (3.6) and the bremsstrahlung spectrum (3.15). Drop the phase and the exact solution in the Appendix displaces photons by J/ω ~ 1/√ω per mode, giving dN/dk ~ k² instead of QED's dN/dk ~ 1/k. The model's cloud is not even the IR-divergent one — it is IR-finite per mode and its total photon number diverges as Λ³, which the authors wave off.

The Section 5 power counting does not fix this. The Gordon-identity estimate δj = O(k/m) bounds the spinor matrix element but misses the phase difference e^{iω_p t} − 1, which at fixed time is O(kt) and after time integration produces 1/(ω_p − ω) ~ 1/(p·k). The loop integral ∫d³k/ω |δj|² ~ ∫k³ dk has no energy denominator, so the estimate concerns a different, softer object than the actual IR-singular integrand. The stress-test note lands; I checked the equations cited.

Credit where due: Section 3 is clean and matches Bloch–Nordsieck and Jauch–Rohrlich. The Appendix's diagonalization of the solvable model is competently done. The L-functional formalism is a legitimate home for states outside Fock space and is the right kind of setting for this problem — but it is the authors' prior work, not a new result here.

Beyond the flaw, the promised diagram technique is never actually delivered: no Feynman rules, no worked diagram, no explicit demonstration that amputated GGreen functions are IR-finite on shell. The physics result — IR-finite inclusive cross sections — is acknowledged as well known in the first sentence, so the technique is the whole show, and it is only a sketch. OCR corruption makes parts hard to parse, a real but secondary annoyance.

Who gets value: a specialist in soft-photon structure or eikonal methods, reading to judge whether the 1/m-around-the-eikonal split can be repaired — the obvious fix is to put the p·k/p⁰ phase into H_as, i.e., actually use Kulish–Faddeev. It deserves a serious referee: the claim is substantive, the authors are credible, and the hole is precise enough to referee cleanly. But as written, the identification of H_as with the IR content of QED fails, so my verdict is reject.

Referee Report

4 major / 5 minor

Summary. The paper proposes a formalism based on L-functionals to eliminate infrared divergences in inclusive QED cross sections. The authors introduce a solvable eikonal Hamiltonian H_as that they claim captures all infrared-divergent structure of QED, solve it exactly in the L-functional framework, and then treat the difference between the full QED Hamiltonian and H_as as a perturbation. They argue that this perturbation theory is infrared finite by a power-counting estimate using the Gordon decomposition. The paper also derives bremsstrahlung formulae and presents an appendix with an exact solution of the solvable model.

Significance. If the central claim were correct, the paper would provide a new formulation of QED in which inclusive cross sections are infrared finite without cancellation of divergences, and it would clarify the role of dressed states in a non-Fock representation. The paper is explicit and self-contained in deriving the solvable model, and it gives concrete formulas for bremsstrahlung and for the photon cloud in the solvable sector. However, the significance is conditional on the identification of H_as as the true eikonal limit of QED, and that identification is not established.

major comments (4)

[Section 4, Eqs. (4.1)/(4.5) vs Section 3, Eq. (3.5)] The solvable Hamiltonian H_as omits the recoil phase e^{iω_p(k)t} that appears in the true current j_free(p,k,t) = -e p^μ/p^0 e^{i(ω_p(k)-ω(k))t} in Eq. (3.5). This phase is precisely what produces the 1/(p·k) denominators in the Liénard–Wiechert field (3.6) and the Bloch–Nordsieck spectrum (3.15). The interaction (4.1)/(4.5) contains a time-independent charge density and photon operators e^{±iω(k)t}, so its photon displacement in Appendix A.2 scales as J/ω ~ 1/√ω, giving a soft-photon number dN/dk ~ const, not the QED dN/dk ~ 1/k. The paper's claim that H_as is the eikonal limit of the true interaction is therefore not justified; H_as is a different, unphysical model with different infrared behavior.
[Section 5, power-counting argument] The estimate δj^μ = O(k/m) from the Gordon identity controls only the spinor matrix element. It does not control the phase factor e^{i(ω_p(k)-ω(k))t} - 1, which is not O(k/m) and is the actual source of the 1/(p·k) poles. Consequently, the loop integral ∫ d^3k/ω |δj|^2 is not just ∫ k^2 dk · k^{-1} · (k/m)^2; it must also include the energy denominator from time integration, which can yield ∫ dk/k. The claimed convergence at k→0 is therefore not established. This is the load-bearing step for the central claim of infrared finiteness.
[Section 5, 'H_as carries all the IR-dangerous structure'] The paper asserts that H_as 'is precisely the eikonal projection of the true current' and that it is 'the unique source of IR divergences in conventional QED perturbation theory.' This is not proven; the text uses heuristic phrases ('One can say', 'We argue') rather than a derivation. Given that the exact solution in Appendix A.2–A.3 produces a photon displacement different from the QED soft-photon cloud, this assertion is exactly what needs to be demonstrated, and the paper does not do so.
[Appendix A.5 and Section 4] The paper itself notes that the fermionic Green function of the solvable Hamiltonian differs from the free one by 'a factor related to dressing of the electron by the photon cloud' and a mass renormalization. But the dressing factor Ψ_p(t) in Eq. (A.5) is computed with the time-independent current J^μ(k) of the solvable model. This dressing is not the same as the QED dressing generated by the Liénard–Wiechert field; therefore the statement that these modifications 'have no bearing on the IR analysis of Section 5' is unsupported. The appendix actually provides evidence that the solvable model's infrared sector is different from QED's.

minor comments (5)

[Abstract and Section 5] The abstract promises 'a new diagram technique where the diagrams ... do not contain infrared divergences,' but the paper does not actually present explicit diagrammatic rules; it only says the technique is 'very similar' to standard GGreen-function diagrams. A clear statement of the new diagrammatic elements would improve readability.
[Section 1, after Eq. (1.11)] There is a typo: 'As in nclusive scattering matrix' should be 'As in inclusive scattering matrix.' Also, the sentence 'we are doing the Fourier transform with respect to spatial coordinates and time coordinates' is redundant.
[Eq. (4.11)] The expression for ΦΛ(t,t0,p1,p2) appears to have unbalanced parentheses and a possibly misplaced term 'Λ√2π (t−t0)'. Please check the formula.
[Section 4] The paper states that the solvable Hamiltonian is 'closely related to the asymptotic Hamiltonian of [6]' but does not specify in what precise sense. A direct comparison with the Kulish–Faddeev Hamiltonian would help the reader assess the validity of the eikonal identification.
[Throughout] Several passages are written in an informal style ('It is easy to check,' 'Obviously'), which is acceptable in a physics paper, but the key claims about infrared finiteness are precisely the ones that need rigorous justification.

Circularity Check

0 steps flagged · score 2.0 of 10

No load-bearing circularity; central derivation is self-contained apart from minor non-load-bearing self-citation.

full rationale

The paper's central move is to split the QED Hamiltonian as H = H_as + δV, solve H_as exactly, and argue that δV is IR-regular because its vertex is O(k/m) by the Gordon identity. This is not circular: the smallness of δV is checked by an independent estimate rather than imposed by definition. The conclusion that the inclusive scattering matrix is IR finite is a derived consequence of that estimate, and the estimate is not fitted to the result. The L-functional formalism is introduced via the authors' own references [1]-[4], but Section 1 also defines the formalism directly, so the self-citation is not load-bearing. The skeptical concern that H_as omits the recoil phase e^{i(ω_p(k)-ω(k))t} of the true current, while Eq. (3.5)-(3.6) show that this phase produces the 1/(p·k) poles, is a substantive correctness risk; if the O(k/m) power counting of Section 5 fails, the argument would be wrong, but it would not be circular, because the paper does not define H_as in terms of the absence of IR divergences in δV. No equation or fitted parameter is shown to be equivalent to its own output by construction. The only mild concern is reliance on the authors' own prior work for the L-functional framework, which is not load-bearing given the self-contained introduction.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the L-functional formalism (prior work of one author), the ability to define the inclusive S-matrix as a limit, and the identification of the solvable eikonal Hamiltonian with the IR-dangerous part of QED. The cutoff Λ is a regulator chosen by hand. No new physical entities are introduced.

free parameters (1)

Soft-photon cutoff Λ
The solvable Hamiltonian (4.1) depends on a cutoff Λ via χΛ(k)=θ(Λ−|k|). The IR analysis is claimed to be Λ-independent, but no explicit Λ-independent limit is shown. Λ is a regulator chosen by hand.

assumptions (5)

standard math Standard CCR/CAR, Weyl algebra, Gordon identity, and Magnus expansion are valid and applicable.
Used throughout Sections 1, 4, 5 and the Appendix without proof.
domain assumption L-functionals provide a complete and well-defined description of QED states, including non-Fock dressed states.
This is the foundation of the paper, introduced in Section 1 and referenced to the authors' prior work [1]-[4]. No independent evidence or formal verification is provided.
domain assumption The inclusive scattering matrix can be defined as a limit of amputated GGreen functions on shell, and the limit of cut-off theories exists.
Stated in Section 1 without a rigorous definition or proof of the limit.
domain assumption The solvable Hamiltonian V_as (4.1)/(4.5) is the eikonal limit of the QED interaction and contains all IR-divergent structure, including the validity of dropping the recoil phase of the true current.
This is the load-bearing assumption of Section 5; the paper does not prove that the time-independent charge density in V_as reproduces the soft-photon denominators of Section 3.
domain assumption IR finiteness order by order in the 1/m expansion implies IR finiteness of the full inclusive scattering matrix.
The 1/m expansion is not proven to be asymptotic or convergent; the inference from order-by-order to all-orders is assumed.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Infrared Problem in Quantum Electrodynamics." pith.science (2026). https://pith.science/paper/5TAFEWJA

@misc{pith2026260800204,
  author       = {Pith},
  title        = {Pith review of: Infrared Problem in Quantum Electrodynamics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5TAFEWJA}},
  note         = {Machine review of arXiv:2608.00204}
}

read the original abstract

It is well known that the inclusive cross section in QED is infrared finite. In the standard diagram techniques this result follows from cancellation of infrared divergences. We construct a new diagram technique where the diagrams for inclusive cross sections (and, more generally, for inclusive scattering matrix) do not contain infrared divergences.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

10 extracted references · 2 linked inside Pith

[1]

New formulation of quantum theory.Dokl

Shvarts, A.S. New formulation of quantum theory.Dokl. Akad. Nauk SSSR(1967)173, 793

1967
[2]

Quantum mechanics and quantum field theory from algebraic and geometric viewpoints, (2024) Berlin: Springer

Schwarz, A. Quantum mechanics and quantum field theory from algebraic and geometric viewpoints, (2024) Berlin: Springer

2024
[3]

Adiabatic definitions of scattering matrix and inclusive scattering matrix

Schwarz, A. Adiabatic definitions of scattering matrix and inclusive scattering matrix. arXiv preprint arXiv:2412.10634, (2024), published in Advances in Theoretical and Mathematical Physics

arXiv 2024
[4]

Geometric Approach to Quantum Theory

Schwarz, A. Geometric Approach to Quantum Theory. L-functionals. arXiv:2607.17566 [hep-th], (2026)

arXiv 2026
[5]

Keldysh, Diagram Technique for Nonequilibrium Processes, Soviet Physics JETP-USSR 20(4), 1018 (1965),[Zh

L. Keldysh, Diagram Technique for Nonequilibrium Processes, Soviet Physics JETP-USSR 20(4), 1018 (1965),[Zh. Eksp. Teor. Fiz. 47, 1515 (1964)]

1965
[6]

P. P. Kulish and L. D. Faddeev, Asymptotic conditions and infrared divergences in quantum electro- dynamics,Theor. Math. Phys.4(1970) 745. [Teor. Mat. Fiz.4,153(1970)]

1970
[7]

Bloch and A

F. Bloch and A. Nordsieck, Note on the Radiation Field of the electron,Phys. Rev.52(1937) 54–59

1937
[8]

Theory of Photons and Electrons.(1955)

Jauch, J.M., and Rohrlich, F. Theory of Photons and Electrons.(1955). Springer Berlin Heidelberg

1955

Show all 10 references

[9]

W. Magnus. On the exponential solution of differential equations for a linear operator. Communica- tions on Pure and Applied Mathematics, VII:649-673, 1954

1954
[10]

Der Strom der Diracschen Elektronentheorie

Gordon, W. Der Strom der Diracschen Elektronentheorie. Z. Physik 50, 630–632 (1928)

1928

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [1]

New formulation of quantum theory.Dokl

Shvarts, A.S. New formulation of quantum theory.Dokl. Akad. Nauk SSSR(1967)173, 793

1967

[2] [2]

Quantum mechanics and quantum field theory from algebraic and geometric viewpoints, (2024) Berlin: Springer

Schwarz, A. Quantum mechanics and quantum field theory from algebraic and geometric viewpoints, (2024) Berlin: Springer

2024

[3] [3]

Adiabatic definitions of scattering matrix and inclusive scattering matrix

Schwarz, A. Adiabatic definitions of scattering matrix and inclusive scattering matrix. arXiv preprint arXiv:2412.10634, (2024), published in Advances in Theoretical and Mathematical Physics

arXiv 2024

[4] [4]

Geometric Approach to Quantum Theory

Schwarz, A. Geometric Approach to Quantum Theory. L-functionals. arXiv:2607.17566 [hep-th], (2026)

arXiv 2026

[5] [5]

Keldysh, Diagram Technique for Nonequilibrium Processes, Soviet Physics JETP-USSR 20(4), 1018 (1965),[Zh

L. Keldysh, Diagram Technique for Nonequilibrium Processes, Soviet Physics JETP-USSR 20(4), 1018 (1965),[Zh. Eksp. Teor. Fiz. 47, 1515 (1964)]

1965

[6] [6]

P. P. Kulish and L. D. Faddeev, Asymptotic conditions and infrared divergences in quantum electro- dynamics,Theor. Math. Phys.4(1970) 745. [Teor. Mat. Fiz.4,153(1970)]

1970

[7] [7]

Bloch and A

F. Bloch and A. Nordsieck, Note on the Radiation Field of the electron,Phys. Rev.52(1937) 54–59

1937

[8] [8]

Theory of Photons and Electrons.(1955)

Jauch, J.M., and Rohrlich, F. Theory of Photons and Electrons.(1955). Springer Berlin Heidelberg

1955