Classical electromagnetism can account for several familiar relationships in thermal radiation without using photons or quantum statistics, according to a new theoretical analysis. But the same framework does not produce the correct blackbody spectrum—the pattern of radiation across frequencies—or determine a finite absolute Stefan-Boltzmann constant, the study says.
The work examines idealized electromagnetic radiation inside a cavity and asks which features of cavity thermal radiation follow from classical electrodynamics alone. It is an analytic model of Maxwell fields and classical field ensembles, with no empirical participant sample reported.
What the classical calculation can recover
One central result is a relationship between energy flux and energy density. In ordinary language, the calculation links how much radiation energy is present to how much energy crosses a surface. The paper presents that relationship as part of the structure already contained in classical electromagnetism and classical statistical mechanics.
For isotropic radiation, the paper uses the familiar pressure relation P = u/3, where u is the energy density. The result is framed for radiation distributed equally in direction, not for every possible anisotropic electromagnetic field.
Combining that pressure relation with thermodynamics gives the energy density a fourth-power dependence on temperature. The coefficient multiplying that dependence, however, remains undetermined by the classical treatment.
The analysis takes two complementary routes. The wave-based calculation solves Maxwell’s equations, breaks the fields into modes, and averages over space and time to remove oscillating cross terms. A more systematic treatment builds a canonical Hamiltonian, imposes gauge constraints, introduces a functional-integral partition function, and calculates field correlations.
Where the framework runs into trouble
The difficulty appears when the calculation tries to assign a finite total energy to the classical continuum. The ensemble analysis finds two simple-harmonic-oscillator contributions for each wave-vector, matching the classical equipartition picture, but the continuum volume and the number of modes are formally infinite. Ultraviolet and infrared regulation—cutoffs that restrict the shortest and longest wavelengths—are needed to make the formal quantities finite.
Those regulators can cancel out of the pressure-to-energy-density ratio, leaving the stated P/u relation independent of the common cutoff dependence. That cancellation does not make the individual pressure and energy-density values finite in the unregulated continuum, however.
The result separates a temperature scaling from a numerical normalization. Classical theory can supply the fourth-power dependence, the authors argue, but not the coefficient needed for a finite, fully normalized blackbody result.
A subtle distinction about energy flow
The paper’s treatment of energy flow also depends on how the radiation is selected. In a thermal-equilibrium ensemble, the average Poynting vector—the quantity that represents electromagnetic energy flow—is identically zero, meaning there is no net energy flow.
A separate aperture calculation selects waves moving outward through an opening in the cavity. For a given frequency, the paper prints the resulting collision flux as c nω /4. Elsewhere, it gives a conflicting overall factor, so the numerical normalization should be treated cautiously even though the broader flux relationship is part of the paper’s central claim.
The field-correlation calculation likewise keeps only the physical transverse modes: divergence-free field components selected by a transverse projector. This constraint is part of the canonical ensemble used in the paper, not an empirical measurement.
A classical boundary, not a new measurement
Taken together, the derivations support the authors’ interpretation that classical electromagnetism and statistical mechanics contain the basic architecture of radiation thermodynamics, while quantum theory is needed for the correct spectrum and absolute Stefan-Boltzmann normalization. The analysis does not claim a nonzero net equilibrium radiation flow, and it does not supply empirical validation.
The document is an arXiv preprint, version 1, dated 24 August 2026. It reports no funding statement in the supplied front matter. Technical details and supporting derivations are said to appear in the appendices.
For quantitative use, the calculation needs a specified finite-volume and cutoff prescription, and the intended flux normalization needs clarification where the text prints conflicting factors. The pressure result is also limited to isotropic radiation rather than arbitrary anisotropic fields.
Paper data and sources
Original title: The Photon Gas in Classical Mechanics: A Statistical-Mechanical Treatment of Classical Field Theory
Authors: Farhang Loran, Saman Moghimi-Araghi
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-24
DOI: 10.30511/ttmp.2026.2096221.1090
Original paper · Full text