Pith. sign in

REVIEW 3 major objections 5 minor 35 references

Bose-Einstein condensates' repulsion skews time-of-flight fits by tens of percent

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-03 10:17 UTC pith:O7RFQL2P

load-bearing objection A practical error map for standard Bose-enhanced fits, built on a simulation whose main simplification (ballistic expansion) is acknowledged but not quantified. the 3 major comments →

arxiv 2601.10415 v1 pith:O7RFQL2P submitted 2026-01-15 cond-mat.quant-gas physics.atom-phquant-ph

Cloud parameter estimation for interacting BEC after time-of-flight

Rasmus Malthe Fiil Andersen , Stine Frederiksen , Laurits Stokholm , Ilja Zebergs , Mick Kristensen , Carrie Weidner , Jan Joachim Arlt This is my paper
classification cond-mat.quant-gas physics.atom-phquant-ph
keywords Bose-Einstein condensationtime-of-flight imagingthermal cloud expansionmean-field interactionsBose-enhanced distributionparameter estimationsemi-ideal modelballistic expansion
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper argues that in partially condensed Bose gases, the repulsive mean-field push of the condensate on the surrounding thermal atoms deforms the expanding cloud, so the standard analytical 'Bose-enhanced' fit used to extract temperature and thermal atom number from time-of-flight images is systematically wrong. The authors simulate the expansion ballistically while including the in-trap repulsion, then fit the simulated images with the standard function. They find thermal atom-number errors up to tens of percent and temperature errors up to about ten percent, and the errors do not disappear at long expansion times. Using the simulation itself as the fitting function on experimental images gives condensed fractions that agree with the semi-ideal model far better than the standard fit.

Core claim

Below the critical temperature, the condensate acts as a repulsive mean-field barrier that pushes thermal atoms out of the trap centre. During time-of-flight this reshapes the thermal cloud: at short times the column density develops a central dip, and at long times it becomes more peaked than the non-interacting Bose-enhanced distribution. Because the standard analytical fit cannot represent either deformation, it over- or under-estimates the thermal atom number and temperature depending on geometry, temperature, and flight time, with errors that persist in the long-time limit. The simulation, when used as a fitting model, reproduces the semi-ideal model's condensed fraction on experimental

What carries the argument

The central object is the effective single-particle potential V_th = V + 2U0 n0 — the mean-field repulsion of a Thomas-Fermi condensate acting on thermal atoms — which the simulation retains for the in-trap cloud while treating the release as ballistic. The expansion profile is then an integral over initial positions of a Bose distribution inside this deformed potential; the integral has no closed form and is evaluated numerically on grids, exploiting cylindrical symmetry. These deformed profiles are what cause fitted parameters to shift.

Load-bearing premise

The load-bearing approximation is that all interactions switch off abruptly when the trap is released, so the expansion is purely ballistic; if mean-field forces during the dense early phase of expansion are non-negligible, the simulated profiles and the error estimates shift.

What would settle it

Take a single partially condensed cloud, image it after two different times of flight, extract temperature and thermal atom number with the Bose-enhanced fit, and check whether the difference between the two extractions matches the simulation's predicted sign reversal and magnitude; alternatively, compare time-of-flight-extracted temperatures with a simultaneous in-situ minimally destructive temperature measurement — a mismatch larger than the predicted errors would falsify the model.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • Thermal atom numbers extracted with the standard Bose-enhanced fit can be wrong by tens of percent, and temperatures by up to about 10%, depending on trap shape and reduced temperature.
  • The errors do not vanish at long flight times, so simply waiting longer before imaging does not make the standard fit accurate.
  • For a given experiment, the simulation can provide a correction map — error versus atom number, temperature, and time of flight — to place systematic uncertainties on the usual analysis.
  • Using the simulated profile as the fit function removes the largest part of the bias and brings extracted condensed fractions into line with the semi-ideal model.
  • There is a trade-off: spherical traps give better temperature estimates but worse thermal-atom-number estimates when the standard fit is used.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If interactions during the early, dense part of time-of-flight were included, the error magnitudes and sign-reversal times would likely shift; the paper's conclusion that errors persist at long times is a lower bound for the realistic effect.
  • The same numerical machinery could be extended to extract other quantities, such as the condensed fraction or chemical potential directly, and to non-axisymmetric traps by dropping the cylindrical-symmetry assumption.
  • A direct experimental test would compare time-of-flight-extracted temperatures with an in-situ, minimally destructive measurement on the same cloud; agreement would validate the simulation as a correction tool.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper addresses systematic errors in extracting temperature and thermal atom number from time-of-flight (TOF) absorption images of partially condensed Bose gases. The authors simulate the expansion of the thermal cloud by sampling the in-trap phase-space distribution in the combined harmonic trap plus the mean-field repulsion of a Thomas-Fermi condensate, then propagating the atoms ballistically to the image plane. Fitting these simulated profiles with the standard non-interacting Bose-enhanced distribution (Eq. (7)) yields error maps for N_th and T as functions of trap geometry, atom number, reduced temperature, and TOF (Figs. 5–9). The simulation is also used directly as a fitting function for experimental images, and the extracted BEC fractions are reported to agree better with the semi-ideal model than do results from the Bose-enhanced fit (Fig. 10). The central quantitative claims are that thermal-atom-number errors can reach tens of percent and temperature errors about 10%, and that these errors do not vanish at long TOF.

Significance. If the simulation is trustworthy, this is a practically useful study: TOF fitting with the Bose-enhanced distribution is ubiquitous, and quantified correction maps for common trap geometries and temperatures would be valuable. The paper has genuine strengths: the forward simulation is not circular in the fitting sense because the input parameters are known; grid convergence is explicitly checked in Figs. 2–3; the algorithm is described in sufficient detail to be reproduced; and the phenomenological explanations of the error patterns are plausible. However, the central results rest on the ballistic-expansion assumption, whose quantitative validity is not established, and the experimental validation is made against a model that shares the same conceptual structure as the simulation. These limitations must be addressed before the error maps can be relied upon as quantitative corrections.

major comments (3)
  1. [Sec. 3, Eq. (10)] The effective potential used in the simulation is written as V_eff = V + 2U_0 n_0 − μ = |μ − V|. The second equality is not correct: inside the Thomas-Fermi condensate V + 2U_0 n_0 = μ, so V_eff should be zero there, whereas |μ − V| is generally nonzero. If the numerical code implements the displayed equality literally, the in-trap distribution inside the condensate is wrong; if the code uses the correct piecewise expression, then Eq. (10) misstates the model and should be corrected. Because Eq. (10) is the foundation for all subsequent error maps, this inconsistency must be resolved and the implementation clarified.
  2. [Sec. 3 and Sec. 6] The load-bearing approximation is that interactions cease abruptly at release: the thermal cloud is propagated ballistically after the trap is switched off. The authors explicitly acknowledge that 'the simulations did not take interactions during the time-of-flight into account' and defer the issue to future work. This is not a minor caveat. At release the condensate density is maximal, and the mean-field force 2U_0|∇n_0| on nearby thermal atoms acts for a time ~1/ω. For typical parameters (μ/k_B ~ 100 nK, R ~ 10 μm), the resulting velocity change over a few ms is comparable to thermal velocities at T ~ 100 nK, so the early-time expansion can differ substantially from the ballistic trajectory. This would systematically shift the simulated profiles, the error maps in Figs. 5–9, and the experimental fits in Fig. 10. The manuscript should either quantify the validity of the ballistic approx
  3. [Sec. 5, Fig. 10] The experimental validation compares the simulation-based fit and the Bose-enhanced fit against the semi-ideal model [21]. This model, like the simulation, is built on Hartree-Fock mean-field repulsion from the condensate and neglects post-release interactions. The agreement in Fig. 10 therefore partly reflects internal consistency between the two models rather than an independent validation of the physical assumptions. The authors do note that the BEC and thermal components are fitted independently, which weakens the circularity, but the benchmark is still not independent. I recommend adding a discussion of this limitation or supplementing the comparison with measurements that do not rely on the same expansion model, such as in-situ probes or known thermodynamic relations.
minor comments (5)
  1. [Throughout] There are several typographical and LaTeX artifacts, including 'for for', 'uni00A0' in axis labels, and placeholder text 'LINK-LINK-LINK-LINK' in reference [13]. These should be cleaned before publication.
  2. [Sec. 2, Eq. (2) and Eq. (7)] The polylogarithm notation g_3/2 and g_2/g_3 is used without defining the subscript convention consistently. In particular, g_3(1) in Eq. (7) should be identified explicitly as ζ(3) or defined as the appropriate polylogarithm value.
  3. [Fig. 10] The text says the uncertainties of the mean values are too small to be visible. Please provide error bars in a table or an inset, since the claimed agreement with the semi-ideal model depends on the size of these uncertainties.
  4. [Sec. 4, Figs. 5–9] The color maps are not reproducible from the printed text alone. It would be helpful to state the exact numerical ranges and to mention whether the stripes visible in Fig. 6 are from the stated <1% numerical oscillation or from the plotting interpolation.
  5. [Sec. 3] The complexity statement O(ν^3 ξ^3) is useful, but the description of the parallel implementation on a GPU is very brief. A sentence on memory usage or typical runtime for the reported grids would aid reproducibility.

Circularity Check

0 steps flagged

No significant circularity: the error estimates are forward Monte Carlo comparisons with known simulation inputs, and the experimental validation is an external consistency check against the semi-ideal model.

full rationale

The central error analysis is a forward simulation: the paper chooses physical parameters (Ns, Ts, trap frequencies), generates a thermal-cloud expansion profile using an explicit model (Eq. (10)), and then fits that simulated profile with the Bose-enhanced distribution Eq. (7), comparing fitted Nth and T to the known simulation inputs. This is an open-loop numerical experiment, not a fit of a parameter that is then renamed as a prediction. The paper explicitly defines the errors as ΔNth = Nth - Nth,s and ΔT = T - Ts, so the 'predictions' are comparisons against known inputs. The experimental section is also not circular: the simulated distribution is used as a fitting function for independent experimental images, and the extracted condensed fraction is compared with the external semi-ideal model of Naraschewski and Stamper-Kurn (Ref. [21]). Although the simulation's in-trap effective potential is closely related to the semi-ideal model's potential, the simulation additionally assumes ballistic time-of-flight and is fitted to data independently; the paper itself notes this: 'Since the BEC and thermal components are fitted independently, the use of the semi-ideal model in the simulation of the cloud expansion does not guarantee that the BEC fraction follows the theoretically expected behavior as a function of temperature.' The stated limitation that 'The simulations did not take interactions during the time-of-flight into account' is an acknowledged approximation affecting the validity of the error estimates, not a circular step. Self-citations (e.g., Refs. [4] and [12]) support imaging calibration and apparatus details and are not load-bearing for the central derivation. No uniqueness theorem, ansatz, or fitted parameter is imported from the authors' prior work to force the result.

Axiom & Free-Parameter Ledger

3 free parameters · 6 axioms · 0 invented entities

The paper introduces no new particles, mediators, forces, or conserved quantities. It uses standard mean-field and Thomas-Fermi physics; the principal free choices are numerical grid sizes, fit-region radii, and the undisclosed initial condensate number in simulations.

free parameters (3)
  • Grid sizes nu and xi = nu=40, xi=60
    Numerical resolution parameters chosen from convergence tests in Figs. 2-3; they affect accuracy but not the physical model.
  • Fit exclusion radii = 1.2 R_i (inner), 3 w_i (outer)
    Hand-chosen boundaries for the thermal-cloud fitting region; the reported error estimates depend on these choices.
  • Initial condensate number N0,s in simulations = Undisclosed
    Given N_s and T_s, the paper does not state how N0,s and Nth,s are split; the computed error maps depend on this split.
axioms (6)
  • domain assumption At release, the thermal cloud is in equilibrium with the Bose-Einstein distribution under the condensate mean-field potential, and all interactions cease for the subsequent expansion.
    This is the central semi-ideal/ballistic modeling assumption, introduced in Sec. 3 and acknowledged in Sec. 6.
  • standard math The condensate is described by the Thomas-Fermi approximation and expands according to the scaling laws in Eq. (5).
    Standard BEC theory from Refs. [15,18,19]; used in Secs. 2.2 and 3.
  • domain assumption The thermal mean-field term 2U0 n_th is negligible in both effective potentials.
    Stated in Sec. 3 for low thermal density; this decouples the BEC and thermal components.
  • domain assumption The cloud and trap possess cylindrical symmetry, allowing a reduction of the numerical grid to two dimensions.
    Stated in Sec. 3; valid for the experimental geometries considered but limits generality.
  • standard math For the analytic Bose-enhanced fit, the fugacity is set to z=1 (chemical potential zero) for T<Tc.
    Standard ideal-gas treatment for T<Tc, discussed in Sec. 2.3.
  • domain assumption The semi-ideal model of Ref. [21] is the correct benchmark for condensed fraction as a function of temperature.
    Used in Sec. 5 to validate the simulation-based fit; because the simulation is itself semi-ideal-like, this is not a fully independent standard.

pith-pipeline@v1.3.0-alltime-deepseek · 10717 in / 15535 out tokens · 155269 ms · 2026-08-03T10:17:48.235697+00:00 · methodology

0 comments
read the original abstract

Experiments on Bose-Einstein condensates at finite temperature typically extract the system parameters, such as temperature, atom number, and condensed fraction from time-of-flight images taken after a free expansion time. This paper systematically examines the effect of repulsive interactions between the condensed and thermal atoms in partially condensed clouds on the expansion profile of the thermal cloud. An analytical expression for the expansion can be obtained only if the interactions between the Bose-Einstein condensate and thermal atoms are neglected, resulting in a Bose-enhanced distribution for the thermal component. Here, the deformation of the cloud due to interactions and the effects on estimated parameters are investigated by simulating the expansion using a ballistic approximation. By fitting the simulated expansion profiles with a Bose-enhanced distribution, the errors of using such a fit are estimated, and the results are explained phenomenologically. The simulation was also used as a fitting function for experimental data, showing better agreement of the extracted condensed fraction with the semi-ideal model than results from a Bose-enhanced fit.

Figures

Figures reproduced from arXiv: 2601.10415 by Carrie Weidner, Ilja Zebergs, Jan Joachim Arlt, Laurits Stokholm, Mick Kristensen, Rasmus Malthe Fiil Andersen, Stine Frederiksen.

Figure 1
Figure 1. Figure 1: Illustration of the calculation of the thermal cloud profile after time-of-flight. The initial cloud, as well as [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Relative errors in thermal atom number ( [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Relative errors in thermal atom number ( [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Thermal column densities in a spherical trap configuration with [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Errors in thermal atom number estimation as a function of reduced temperature [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Errors in temperature estimation as a function of reduced temperature [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Errors in relative thermal atom numbers and temperature for different temperatures. The simulations show [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Relative errors in thermal atom number ( [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Errors in relative thermal atom number ( [PITH_FULL_IMAGE:figures/full_fig_p010_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: BEC fraction as a function of reduced temperature [PITH_FULL_IMAGE:figures/full_fig_p010_10.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

35 extracted references · 13 canonical work pages

  1. [1]

    Ketterle W, Durfee D S and Stamper-Kurn D 1999 Making, probing and understanding Bose-Einstein condensates (IOS Press) pp 67–176

    1999

  2. [2]

    Reinaudi G, Lahaye T, Wang Z and Guéry-Odelin D 2007Optics letters323143–3145

  3. [3]

    Veyron R, Mancois V, Gerent J B, Baclet G, Bouyer P and Bernon S 2022Physical Review Research4033033

  4. [4]

    Vibel T, Christensen M B, Kristensen M A, Thuesen J J, Stokholm L N, Weidner C A and Arlt J J 2024Journal of Physics B: Atomic, Molecular and Optical Physics

  5. [5]

    Hueck K, Luick N, Sobirey L, Siegl J, Lompe T, Moritz H, Clark L W and Chin C 2017Optics express258670–8679

  6. [6]

    Szczepkowski J, Gartman R, Witkowski M, Tracewski L, Zawada M and Gawlik W 2009 Review of Scientific Instruments80

    2009

  7. [7]

    Muessel W, Strobel H, Joos M, Nicklas E, Stroescu I, Tomkovič J, Hume D B and Oberthaler M K 2013Applied Physics B11369–73

  8. [8]

    Genkina D, Aycock L, Stuhl B, Lu H I, Williams R and Spielman I 2015New Journal of Physics18013001

  9. [9]

    Putra A, Campbell D L, Price R M, De S and Spielman I 2014Review of Scientific Instruments85

  10. [10]

    Ramanathan A, Muniz S R, Wright K C, Anderson R P, Phillips W D, Helmerson K and Campbell G K 2012Rev. Sci. Instrum.83083119 URLhttps://pubs.aip.org/aip/rsi/ article/83/8/083119/358919/Partial-transfer-absorption-imaging-A-versatile

  11. [11]

    Arora P, Purnapatra S B, Acharya A, Kumar R and Sen Gupta A 2012MAPAN2731–39 ISSN 0974-9853 URLhttps://doi.org/10.1007/s12647-012-0003-3

    work page doi:10.1007/s12647-012-0003-3

  12. [12]

    Vibel T, Christensen M B, Andersen R M F, Stokholm L N, Pawłowski K, Rzążewski K, Kristensen M A and Arlt J J 2024Journal of Physics B: Atomic, Molecular and Optical Physics57195301 URLhttps://dx.doi.org/10.1088/1361-6455/ad7458

    work page doi:10.1088/1361-6455/ad7458

  13. [13]

    thesis Aarhus University URLLINK-LINK-LINK-LINK-LINK-LINK-LINK-LINK

    Andersen M 2025Spectrally Probing Ultra-Cold Bosonic CloudsPh.D. thesis Aarhus University URLLINK-LINK-LINK-LINK-LINK-LINK-LINK-LINK

  14. [14]

    Schroeder D V 2014An Introduction to Thermal Physics1st ed (Pearson)

  15. [15]

    Pethick C J and Smith H 2008Bose-Einstein condensation in dilute gases2nd ed (Cambridge)

  16. [16]

    Gross E P 1961Il Nuovo Cimento20454–477

  17. [17]

    Pitaevskii L P 1961Soviet Physics JETP13451–454

  18. [18]

    Castin Y and Dum R 1996Phys. Rev. Lett.775315

  19. [19]

    Kagan Y, Surkov E L and Shlyapnikov G V 1996Phys. Rev. A54(3) R1753–R1756 URL https://link.aps.org/doi/10.1103/PhysRevA.54.R1753

    doi:10.1103/physreva.54.r1753

  20. [20]

    Bagnato V, Pritchard D E and Kleppner D 1987Phys. Rev. A35(10) 4354–4358 URL https://link.aps.org/doi/10.1103/PhysRevA.35.4354 11

    work page doi:10.1103/physreva.35.4354

  21. [21]

    Naraschewski M and Stamper-Kurn D M 1998Phys. Rev. A582423

  22. [22]

    Giorgini S, Pitaevskii L P and Stringari S 1997Phys. Rev. Lett.783987

  23. [23]

    Griffin A 1996Phys. Rev. B539341

  24. [24]

    Görlitz A, Vogels J M, Leanhardt A E, Raman C, Gustavson T L, Abo-Shaeer J R, Chikkatur A P, Gupta S, Inouye S, Rosenband T and Ketterle W 2001Phys. Rev. Lett.87(13) 130402 URLhttps://link.aps.org/doi/10.1103/PhysRevLett.87.130402

    work page doi:10.1103/physrevlett.87.130402

  25. [25]

    Schweikhard V, Coddington I, Engels P, Mogendorff V P and Cornell E A 2004Phys. Rev. Lett.92(4) 040404 URLhttps://link.aps.org/doi/10.1103/PhysRevLett.92.040404

    doi:10.1103/physrevlett.92.040404

  26. [26]

    Burger S, Cataliotti F S, Fort C, Maddaloni P, Minardi F and Inguscio M 2002Europhysics Letters571 URLhttps://doi.org/10.1209/epl/i2002-00532-1

    work page doi:10.1209/epl/i2002-00532-1

  27. [27]

    Hadzibabic Z, Stock S, Battelier B, Bretin V and Dalibard J 2004Phys. Rev. Lett.93(18) 180403 URLhttps://link.aps.org/doi/10.1103/PhysRevLett.93.180403

    work page doi:10.1103/physrevlett.93.180403

  28. [28]

    Andrews M R, Kurn D M, Miesner H J, Durfee D S, Townsend C G, Inouye S and Ketterle W 1997Phys. Rev. Lett.79(4) 553–556 URL https://link.aps.org/doi/10.1103/PhysRevLett.79.553

    work page doi:10.1103/physrevlett.79.553

  29. [29]

    Ferioli G, Pancaldi S, Glicenstein A, Clément D, Browaeys A and Ferrier-Barbut I 2024Phys. Rev. Lett.132(13) 133601 URL https://link.aps.org/doi/10.1103/PhysRevLett.132.133601

    work page doi:10.1103/physrevlett.132.133601

  30. [30]

    Rychtarik D, Engeser B, Nägerl H C and Grimm R 2004Phys. Rev. Lett.92(17) 173003 URL https://link.aps.org/doi/10.1103/PhysRevLett.92.173003

    work page doi:10.1103/physrevlett.92.173003

  31. [31]

    Hammes M, Rychtarik D, Nägerl H C and Grimm R 2002Phys. Rev. A66(5) 051401 URL https://link.aps.org/doi/10.1103/PhysRevA.66.051401

    work page doi:10.1103/physreva.66.051401

  32. [32]

    Smith N L, Heathcote W H, Hechenblaikner G, Nugent E and Foot C J 2005Journal of Physics B: Atomic, Molecular and Optical Physics38223 URL https://doi.org/10.1088/0953-4075/38/3/007

    work page doi:10.1088/0953-4075/38/3/007

  33. [33]

    thesis Aarhus University

    Christensen M B 2020Microcanonical Fluctuations in Interacting Bose-Einstein Condensates Ph.D. thesis Aarhus University

  34. [34]

    Gajdacz M, Pedersen P L, Mørch T, Hilliard A J, Arlt J and Sherson J F 2013Review of Scientific Instruments84083105 ISSN 0034-6748 (Preprinthttps://pubs.aip.org/aip/ rsi/article-pdf/doi/10.1063/1.4818913/14781030/083105_1_online.pdf) URL https://doi.org/10.1063/1.4818913

    work page doi:10.1063/1.4818913/14781030/083105_1_online.pdf

  35. [35]

    Kristensen M A, Gajdacz M, Pedersen P L, Klempt C, Sherson J F, Arlt J J and Hilliard A J 2017Journal of Physics B: Atomic, Molecular and Optical Physics50034004 URL https://dx.doi.org/10.1088/1361-6455/50/3/034004 12

    work page doi:10.1088/1361-6455/50/3/034004