Angular redistribution of Bose-stimulated light scattering in realistic degenerate Bose gases: interactions, traps, and the low-heating small-angle probe

For a cold ideal Bose gas, bosonic final-state stimulation enhances light scattering at small momentum transfer, and the enhancement grows dramatically below the Bose–Einstein condensation temperature $T_c$ while shifting to ever smaller scattering angles. Politzer (1995) proposed exploiting this by collecting scattered light at small angles, where the stimulated channel carries signal without heating the gas, enabling an essentially non-destructive probe. That prediction was derived for a uniform, non-interacting gas, and the source paper itself flags the simplifications — atomic interactions, finite trap size and non-conserved momentum, diffraction limits on accessible angles, free-expansion dynamics, and interplay with coherence and line-shape effects — whose relaxation decides experimental feasibility. Prior work: the ideal-gas trapped problem was solved analytically by Politzer (1997) and generalized in 2023; the enhancement itself was first observed by Lu, Margalit and Ketterle (2023) using a high-density gas at a fixed large scattering angle; and the quantitative effect of atomic interactions was measured near $T_c$ by Konstantinou et al. (2026), who found that weak repulsive interactions almost completely suppress the enhancement while attractive ones enhance it. What remains open is the angle-resolved problem: a quantitative, interaction-inclusive theory and an experimental measurement of the full angular (momentum-transfer) distribution of the Bose-stimulated scattering rate above and below $T_c$ — in particular whether the dramatic ideal-gas small-angle enhancement survives interactions in the condensed regime — together with a demonstration (or refutation) of the small-angle, low-heating probing strategy.

Background

When a photon is elastically scattered by an atom, the atom recoils and changes its motional quantum state. For identical bosons, the transition rate into a final state already occupied by $N_f$ atoms is enhanced by the factor $N_f+1$ — bosonic stimulation, the matter-wave analog of stimulated emission of light. In a dilute gas cooled toward the Bose–Einstein condensation temperature $T_c$ (below which a macroscopic fraction of atoms accumulates in the single-particle ground state), the low-momentum motional states become highly occupied, so light scattering that transfers only a small momentum $q$ to the atoms is strongly enhanced: the final recoil state is likely to be occupied already. Politzer (1991) and Javanainen (1994) were among the first to analyze light scattering from Bose-condensed gases; Javanainen's treatment of a tightly confined condensate showed that in a trap the atomic momentum is not conserved and ground-state-to-ground-state recoilless transitions become analogs of the Mössbauer effect (the nuclear-physics phenomenon where a crystal as a whole absorbs recoil, sharpening gamma-ray emission lines).

Politzer (1995) computed, in the Born approximation (single-scattering, weak probe — appropriate for an optically thin sample) for a uniform, ideal (non-interacting) Bose gas, the total scattering rate per unit solid angle, normalized to ordinary Rayleigh scattering (the isotropic, unstimulated background), as a universal function $R(\Delta,\tau)$ of the dimensionless momentum transfer $\Delta$ (photon momentum transfer in units of the thermal momentum at $T_c$ ) and reduced temperature $\tau=T/T_c$ . The rate splits into three channels: the isotropic Rayleigh term; stimulated scattering between thermally occupied nonzero-momentum states (a term involving a one-dimensional integral that must be evaluated numerically); and, below $T_c$ , stimulated scattering between the condensate and thermal states. Above $T_c$ the enhancement at $\Delta\lesssim 1$ is a factor of a few; below $T_c$ it can reach several hundred to a thousand for $\Delta\lesssim 0.1$ , concentrating at progressively smaller angles as $T$ drops. Because the stimulated channel moves atoms within the thermal distribution, it adds essentially no net energy to the gas, so detecting at milliradian angles could in principle cut the incident intensity — and hence the heating — by the enhancement factor for the same signal. The paper notes explicitly that the analysis assumes a uniform, ideal gas; that diffraction from a finite sample limits the smallest meaningful angles, placing the interesting regime at the border of observability for then-current experiments; that in a real trap momentum is not conserved and the Mössbauer-like transitions acquire a width set by the ground-state momentum spread; and that under free expansion after trap release the momentum-space overlaps that carry the enhancement decay, confining the signal to still smaller angles.

Politzer (1997) then solved the ideal-gas harmonic-trap problem at large atom number $N$ : comparing Rayleigh scattering, coherent diffraction (the trap analog of the Mössbauer transitions, proportional to the square of the Fourier transform of the density), and the Bose-stimulated channels, with explicit scaling laws — e.g. the condensate-to-excited-state stimulated rate grows like $T/\delta^2$ over a range of trap momentum transfers $\delta$ that widens with $N$ , while only $O(N_e^{1/3})$ of the $N_e$ thermal atoms effectively participate. He left detailed trap-specific numerical evaluations to experiment-specific future work. Two decades later, Lu, Margalit and Ketterle (2023) reported the first observation of bosonic enhancement of light scattering, realizing the small-recoil regime not at small laboratory angles but at 90-degree detection with an extremely dense sodium gas ( $T_c\approx 10~\mu\mathrm{K}$ ); they derived new ideal-gas limits for trapped gases (a universal bound $\zeta(2)/\zeta(3)\approx 1.37$ , with $\zeta$ the Riemann zeta function, on the enhancement above $T_c$ in a harmonic trap, diverging as $1/\kappa^2$ below it, with $\kappa$ the atomic recoil energy — the kinetic energy an atom gains by absorbing one photon — in units of $k_BT_c$ ), discussed interaction corrections, and recommended future measurements at lower densities and smaller scattering angles. Konstantinou et al. (2026) then measured, in a quasi-homogeneous box-trapped potassium gas (a flat-bottomed optical trap realizing a nearly uniform gas) near $T_c$ at a fixed 35-degree angle with tunable interactions, that even weak repulsive interactions (s-wave scattering length $a$ , the parameter setting the strength of low-energy two-body contact interactions, much smaller than the thermal de Broglie wavelength $\lambda$ , the size of the atomic wave packets) almost completely suppress the bosonic enhancement — effectively $N_f\to N_f(1-c\,a/\lambda)$ with $c\approx 12$ at $T_c$ , varying across the experimental density range — while attractive interactions enhance it; the normalized scattering rate (the enhancement factor relative to the single-particle rate) equals the static structure factor $S(q)$ , i.e. the Fourier transform of the second-order density correlation function, so the effect reflects beyond-mean-field local correlations. Both works measure at a single fixed angle and defer angle-resolved measurements to the future. A separate, adjacent line — Bragg spectroscopy (Stenger et al. 1999), where two laser beams stimulate scattering and measure the dynamic structure factor (the frequency-resolved density response) of an interacting condensate — covers the condensed, small-momentum-transfer regime in a different (driven, two-photon) channel, but has not been connected to the spontaneous single-beam angular redistribution of the source question.

Problem Statement

Determine, quantitatively and experimentally, the angular redistribution of Bose-stimulated light scattering in a realistic degenerate Bose gas, and decide from it whether collecting scattered light at small momentum transfer can serve as a low-heating, essentially non-destructive probe. For a specified trapped gas (harmonic or box geometry) with controllable atom number, density, interaction strength, and temperature, this requires three jointly linked results: (1) a quantitative theory of the differential scattering rate per unit solid angle as a function of momentum transfer $q$ and temperature $T$ across the Bose-Einstein transition, including the effect of atomic interactions on the pair correlations that produce the enhancement, which must reduce in limiting cases to the known ideal-gas results — the uniform-gas rate function $R(\Delta,\tau)$ of Politzer (1995) with its three channels, the large- $N$ harmonic-trap scalings of Politzer (1997), and the trapped-gas limits of Lu et al. (2023) ( $\zeta(2)/\zeta(3)$ bound above $T_c$ , $1/\kappa^2$ divergence below) — and must be consistent, in the near-critical quasi-homogeneous regime probed at a fixed scattering angle, with the interaction suppression and enhancement measured by Konstantinou et al. (2026), whose effective coefficient $c\approx 12$ in $N_f\to N_f(1-c\,a/\lambda)$ applies at $T_c$ and varies across their experimental density range; (2) an angle-resolved measurement of the same map — scattering rate versus momentum transfer at several angles spanning the enhancement region, above and below $T_c$ — establishing or refuting the predicted concentration of the enhancement at small momentum transfer below $T_c$ and its interaction-induced suppression; and (3) an assessment, for a realistic geometry including diffraction from the finite sample and beam, of the ratio of the stimulated small-angle signal to the always-present isotropic Rayleigh channel — i.e. the heating reduction actually achievable at a given detected signal strength. These deliverables serve one common objective — an experimentally validated account of the Bose-stimulated small-angle scattering channel sufficient to decide the feasibility of the low-heating probe — so an answer addressing only the ideal gas or only a single fixed angle does not resolve the problem. A valid completion may alternatively show, by the same interaction-inclusive calculation or angle-resolved measurement, that the enhancement window is destroyed by interactions or made inaccessible by diffraction throughout the accessible range of the specified gas's controllable parameters (atom number, density, interaction strength, temperature) and probed momentum transfers, refuting the strategy. Such a refutation resolves the task on its own, without additionally requiring the other deliverables. Related directions discussed in this lineage — the behavior under time-dependent protocols such as free expansion after trap release, and the integration of coherence, line-shape, pulsed-light, and other optical many-body effects into the same predictive model — remain of interest but are not required to resolve the problem.

References

H. David Politzer, Optical scattering angles and Bose-stimulated motion in cold atomic gas, Physics Letters A 209, 160-164 (1995). DOI: 10.1016/0375-9601(95)00812-7; arXiv:quant-ph/9507012; https://arxiv.org/abs/quant-ph/9507012
H. David Politzer, Bose-stimulated scattering off a cold atom trap, Physical Review A 55, 1140-1146 (1997). DOI: 10.1103/PhysRevA.55.1140; arXiv:quant-ph/9608044; https://arxiv.org/abs/quant-ph/9608044
Juha Javanainen, Optical signatures of a tightly confined Bose condensate, Physical Review Letters 72, 2375-2378 (1994). DOI: 10.1103/PhysRevLett.72.2375; https://doi.org/10.1103/PhysRevLett.72.2375
H. David Politzer, Light incident on a Bose-condensed gas, Physical Review A 43, 6444-6446 (1991). DOI: 10.1103/PhysRevA.43.6444; https://doi.org/10.1103/PhysRevA.43.6444
Yu-Kun Lu, Yair Margalit, Wolfgang Ketterle, Bosonic stimulation of atom–light scattering in an ultracold gas, Nature Physics 19, 210 (2023; published online 22 December 2022; preprint title: Observation of bosonic stimulation in light scattering). DOI: 10.1038/s41567-022-01846-y; arXiv:2204.06639; https://arxiv.org/abs/2204.06639
Konstantinos Konstantinou, Yansheng Zhang, Paul H. C. Wong, Feiyang Wang, Yu-Kun Lu, Nishant Dogra, Christoph Eigen, Tanish Satoor, Wolfgang Ketterle, Zoran Hadzibabic, Suppression and enhancement of bosonic stimulation by atomic interactions, Nature Physics 22, 362-366 (2026). DOI: 10.1038/s41567-025-03155-6; arXiv:2503.24386; https://arxiv.org/abs/2503.24386
J. Stenger, S. Inouye, A. P. Chikkatur, D. M. Stamper-Kurn, D. E. Pritchard, W. Ketterle, Bragg spectroscopy of a Bose-Einstein condensate, Physical Review Letters 82, 4569-4573 (1999). DOI: 10.1103/PhysRevLett.82.4569; arXiv:cond-mat/9901109; https://arxiv.org/abs/cond-mat/9901109
Weiping Zhang, C. A. Sackett, R. G. Hulet, Optical detection of a Bardeen-Cooper-Schrieffer phase transition in a trapped gas of fermionic atoms, Physical Review A 60, 504-507 (1999). DOI: 10.1103/PhysRevA.60.504; https://doi.org/10.1103/PhysRevA.60.504
D. Meiser, P. Meystre, Reconstruction of the phase of matter-wave fields using a momentum-resolved cross-correlation technique, Physical Review A 72, 023605 (2005). DOI: 10.1103/PhysRevA.72.023605; arXiv:cond-mat/0502625; https://arxiv.org/abs/cond-mat/0502625
J. H. Müller, P. Petrov, D. Oblak, C. L. Garrido Alzar, S. R. de Echaniz, E. S. Polzik, Diffraction effects on light-atomic-ensemble quantum interface, Physical Review A 71, 033803 (2005). DOI: 10.1103/PhysRevA.71.033803; https://doi.org/10.1103/PhysRevA.71.033803
R. M. F. Andersen, L. N. Stokholm, I. Zebergs, N. R. Neubert, A. S. Chatterley, N. Kjaergaard, J. J. Arlt, Optical spectroscopy of Bose-Einstein condensates at finite temperature (2026). arXiv:2606.24568; https://arxiv.org/abs/2606.24568
Amilson R. Fritsch, Hector Letellier, Leonardo Lima da Silva, Pierre Azam, Vanderlei Salvador Bagnato, Robin Kaiser, Mathilde Hugbart, In situ coherence measurements of scattered light in magnetically trapped cold atomic clouds: Probe-driven atomic dynamics (2026). arXiv:2607.18808; https://arxiv.org/abs/2607.18808