REVIEW 3 major objections 5 minor 32 references
Observation of the algebraic localization-delocalization transition in a 1D disordered potential with a bias force
T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read A single dimensionless parameter α governs the localization–delocalization transition of ultracold atoms in a one-dimensional disordered potential with a bias force, with the transition observed at α = 1.0(3).
desk verdict First observation of the biased-force algebraic localization-delocalization transition, with a convincing alpha collapse but an under-tested universality claim due to fixed dimensionless time. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the dimensionless control parameter α = ℏ²a/C̃(0), which compares the energy gained from the bias force over a localization length to the disorder energy scale; the transition to delocalization is predicted at α = 1. The velocity-dependent variant α* = ℏ²a/Ĉ(2k) incorporates the speckle power spectrum at the atomic wavevector, which is how the initial velocity enters. The experimental machinery consists of a potassium-39 condensate launched into a one-dimensional tube, a speckle potential with measured power spectral density, and a magnetic-field-gradient force; the measured quantity is the localized atomic fraction extracted by integrating density near the launch position. The claim that α is sufficient is carried by the data collapse onto 1/√α and by numerical simulations of the one-dimensional Schrödinger equation used to calibrate the disorder amplitude.
What would settle it
A decisive test would be to measure the localized fraction as a function of 1/√α using an independent, absolute calibration of the speckle amplitude—for example, by directly probing the disorder potential via Bragg spectroscopy or by measuring the localization length at zero force—and to check whether the collapse and the crossing point at α = 1.0(3) remain unchanged. If a second method gives a different transition point or shows that the localized fraction depends on the initial velocity beyond the rescaling by Ĉ(2k), the claim that α is the only relevant parameter would be refuted. Another falsifier is to extend the propagation time well beyond the transient window and observe whether the algebraic decay becomes exponential, which would indicate that the observed transition is a transient phenomenon rather than a true algebraic localization–delocalization transition.
Extended reading notes
Core claim
The paper's core discovery is that, for a one-dimensional disordered potential realized by an optical speckle and a constant bias force, the fraction of atoms that remain localized is a universal function of the single dimensionless parameter α = ℏ²a/C̃(0), where a is the acceleration from the force and C̃(0) is the zero-momentum power spectral density of the disorder. The data for four different accelerations collapse onto one curve when plotted against 1/√α, and the half-localization crossing point is found at α = 1.0(3), in agreement with theoretical predictions. When atoms enter the disorder with a nonzero initial velocity v, the same universality is recovered by replacing C̃(0) with the power spectrum evaluated at the atomic wavevector, Ĉ(2k), giving α*; the transition then occurs at α* = 1.0(4). This demonstrates that the localization–delocalization transition is energy-independent and that the initial velocity enters only through the correlation of the disorder. The paper also shows that the localized density profile is algebraic, with an exponent β that increases with α and saturates for strong disorder, and interprets the saturation and the discrepancy with white-noise analytics as consequences of the correlated, finite-width speckle spectrum.
Load-bearing premise
The analysis assumes that the speckle potential is correctly described as a one-dimensional correlated disorder with the measured power spectral density C(k), and that the disorder amplitude VR is calibrated from the optical power using one-dimensional Schrödinger simulations with a 15% uncertainty; if either the one-dimensional reduction or the VR calibration is biased, the absolute values of α and the inferred transition point would shift, although the data collapse on 1/√α would likely survive.
Editorial extensions
If this is right
- The localized fraction of a non-interacting wave packet in a 1D disordered potential with a bias force depends only on the dimensionless combination α = ℏ²a/C̃(0), so experiments at different forces and disorder strengths can be compared on a single universal curve.
- The insensitivity of the transition to the initial velocity means that the location of the transition point is energy-independent, up to the correlation-induced rescaling by the power spectrum at the atomic wavevector.
- The algebraic decay of the localized density, with an exponent that saturates near 2 for strong disorder, implies that the long-distance tail of the localized wave function in a correlated speckle is not exponential in the observed time window.
- The measured transition at α = 1.0(3) provides a quantitative benchmark against which analytic theories of localization with a bias force, including white-noise and correlated models, can be tested.
- The demonstration that the speckle correlation enters through Ĉ(2k) means that future experiments can use this system to directly probe the momentum-dependent power spectrum of a disordered potential.
Reading between the lines
- If α is truly the only relevant parameter, then the same scaling should hold for other measures of localization, such as the inverse participation ratio or the exponential decay length, a prediction the paper does not test but which could be checked in the same dataset.
- The observed saturation of the algebraic exponent near 2 at strong disorder suggests a possible universal connection to the kc-edge localization of speckle potentials; this could be probed by varying the speckle correlation length σ independently of the amplitude.
- The experiment opens a direct route to study the interplay of interactions and disorder in the presence of a force: since the non-interacting transition is now characterized, tuning the scattering length near a Feshbach resonance could test whether interactions shift the critical α or destroy algebraic localization.
- The data collapse could be used to extract a master curve for the localized fraction versus 1/√α; if such a master curve is indeed universal, it would also apply to other 1D experiments with different atomic species and speckle geometries.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports experiments with a non-interacting 39K Bose-Einstein condensate launched into a 1D laser-speckle disorder under a constant bias force. The authors measure the localized atomic fraction after a finite propagation time for four accelerations and for several initial velocities. They find that the measured localized fractions collapse when plotted against 1/sqrt(alpha), where alpha = hbar^2 a / Ctilde(0), and define the transition at a localized fraction of 0.5, obtaining alpha = 1.0(3) (and alpha* = 1.0(4) when the velocity-dependent effective disorder is used). They also observe algebraic tails in the localized density profiles and extract the decay exponent beta as a function of alpha. The experimental results are compared with disorder-averaged numerical solutions of the 1D Schrödinger equation, which are used both for comparison and to calibrate the speckle amplitude V_R.
Significance. If the central claim holds, this is the first experimental observation of the algebraic localization-delocalization transition in a 1D disordered potential with a bias force, a phenomenon that extends Anderson localization. The collapse of four acceleration curves on a single dimensionless parameter and the velocity rescaling through the power spectrum are clean, falsifiable observations. The paper also carefully averages over disorder realizations and explicitly compares with 1D Schrödinger simulations. However, the 'only relevant parameter' claim is currently underdetermined because the data are all taken at a nearly fixed ratio of the quantum energy scale E_t = hbar/tau to the acceleration energy E_a, and the absolute calibration of V_R relies on the same simulations used for comparison.
major comments (3)
- [Energy scales and Fig. 2(b)] The data collapse is shown at a single dimensionless time: the propagation times tau = 460, 320, 280, 90 ms are chosen so that E_t is below E_a by a factor of the order of 15 to 20 for every acceleration. Since the localized fraction is an escape fraction after a finite time, and the paper itself states that for correlated speckle full delocalization is expected at infinite time with only transient signatures observable, the collapse on 1/sqrt(alpha) is equally consistent with a dependence on both alpha and E_t/E_a evaluated at one nearly fixed ratio. The statement that alpha is the only relevant parameter therefore requires a time-dependence check. I request either an experimental scan of tau at fixed alpha (e.g., two values of E_t/E_a near the transition) or a numerical demonstration that the localized fraction at the reported times is already at its tau-independent value; without this, the transition at alpha = 1.0(3) remains a time-dependent crossover.
- [Calibration of V_R (simulation paragraph)] The horizontal axis in Fig. 2(b) is proportional to V_R, and the definitions alpha = (E_a/V_*)^{3/2} = hbar^2 a / Ctilde(0) with Ctilde(0) = c V_R^2 pi sigma imply alpha is proportional to V_R^{-2}. The stated 15% calibration uncertainty in V_R therefore translates into roughly 30% uncertainty in alpha, which is comparable to the quoted transition value 1.0(3). As written, the agreement with the predicted alpha = 1 is not established beyond the calibration uncertainty. The authors should either propagate the calibration uncertainty into the reported transition point or provide an independent calibration of V_R.
- [Velocity scaling, Fig. 3] The velocity-rescaling result is presented for a single acceleration (a = 19.2 mm/s^2) and a single propagation time (tau = 280 ms). The same finite-time limitation as in the zero-velocity case applies, so the collapse on alpha* demonstrates consistency at that working point but does not by itself establish that alpha* is the only relevant parameter for all times. A sentence stating this limitation, together with the requested tau-dependence check, would make the claim appropriately precise.
minor comments (5)
- [Fig. 2 caption] The caption contains a typo: 'background substraction' should be 'background subtraction'.
- [Experimental sequence] The sentence 'The evaporation is then pursed in the |F=1,mF=1> state' appears to contain a typo; 'pursed' should likely be 'performed' or 'pursued'.
- [Reference [27]] The author list in reference [27] is incomplete and inconsistent ('L. Fouch' should presumably be 'L. Fouché'), and the journal title is abbreviated; please check the reference.
- [Fig. 5 caption] The values of beta are shown as a function of 1/sqrt(alpha) but the caption does not state how the error bars (if shown) are obtained; the text mentions an error estimate from reduced data sets for one case only. Please specify the uncertainty procedure for all plotted points.
- [Operational transition definition] The localized-fraction criterion of 0.5 is an operational choice; since the localized fraction is time dependent, the inferred transition point will depend on this choice. The authors may wish to state explicitly that this is a definition used to compare with theory, not a thermodynamic transition point.
Circularity Check
Transition point α=1.0(3) is inherited from the simulation used to calibrate VR; the data collapse itself is independent.
-
fitted input called prediction
[Page 3, definition of α and discussion of Fig. 2]
"The non-dimensional parameter α is then defined as α = (Ea/V∗)3/2 = ℏ2a/ C(0) ... We use these simulations to calibrate VR as a function of the optical power with a 15% uncertainty. ... If the localized-delocalized transition point is defined for a localized fraction equal to 0.5, this corresponds to α = 1.0(3)"
The α scale is set by VR because α ∝ 1/VR^2 via C(0)∝VR^2. The paper calibrates VR by matching the experimental localized-fraction curves to 1D Schrödinger simulations of the same model. Since this calibration is a global rescaling of VR, the experimental crossing of the localized fraction at 0.5 is forced to occur at the same VR as the simulation's crossing. Hence the reported transition α=1.0(3) is not an independent measurement of the theoretical prediction α=1; it is the simulation's transition point transferred onto the experiment. The collapse on 1/√α is independent of this calibration, but the quantitative transition point is not.
full rationale
The paper's central claim that the localized fraction collapses on 1/√α is an empirical observation not forced by the calibration, because a single global conversion from optical power to VR cannot produce collapse across four accelerations unless the data already scale. However, the quantitative transition point α=1.0(3) is not an independent test of the theoretical α=1: VR, which sets the α scale, is calibrated using the same 1D Schrödinger simulations that embody the model under test, so the crossing value is inherited from the simulation. No self-citation is load-bearing, as the theory references are external. The fixed E_t/E_a choice is a legitimate scientific limitation affecting the claim that α is the only relevant parameter, but it is not a circularity.
Assumptions & free parameters
free parameters (1)
- VR calibration scale (speckle amplitude vs optical power) =
calibrated via 1D Schrödinger simulations, 15% uncertainty
assumptions (5)
- domain assumption The atoms are non-interacting, with scattering length -0.2 +/- 0.2 a0.
- domain assumption The speckle potential is a stationary random potential fully described by its measured power spectral density C(k).
- domain assumption The disorder can be treated as one-dimensional for propagation along x.
- domain assumption The 1D Schrödinger equation with the measured speckle correlations is a valid model for calibrating VR.
- domain assumption Averaging over 8 disorder realizations gives representative localized fractions.
Cite this review
Pith. "Pith review of Observation of the algebraic localization-delocalization transition in a 1D disordered potential with a bias force." pith.science (2026). https://pith.science/paper/4IOVSRCU
@misc{pith2026190801511,
author = {Pith},
title = {Pith review of: Observation of the algebraic localization-delocalization transition in a 1D disordered potential with a bias force},
year = {2026},
howpublished = {\url{https://pith.science/paper/4IOVSRCU}},
note = {Machine review of arXiv:1908.01511}
}
abstract
In a one-dimensional (1D) disordered potential, quantum interferences leading to Anderson lo-calization are ubiquitous, such that all wave-functions are exponentially localized. Moreover, no phase transition toward delocalization is expected in 1D. This behavior is strongly modified in the presence of a bias force. We experimentally study this case, launching a non-interacting 39 K Bose-Einstein condensate in a 1D disordered potential induced by a far-off-resonance laser speckle, while controlling a bias force. In agreement with theoretical predictions, we observe a transition between algebraic localization and delocalization as a function of our control parameter that is the relative strength of the disorder against the bias force. We also demonstrate that the initial velocity of the wave-packet only plays a role through an effective disorder strength due to the correlation of the disorder. Adding a bias force is a quite natural way to probe the transport properties of quantum systems, a subject of broad interest that can be in particular addressed with atomic quantum gases thanks to their high degree of control and versatility [1]. For example, Bloch oscillations has been measured through the addition of a constant force to atoms in periodic potential induced by an optical lattice [2]. A force applied to a harmonic trap is equivalent to a trap displacement. The response to such a displacement permits to reveal the fluid or insulating behavior of atomic systems. In 1D interacting Bose gases, the pinning transition by an optical lattice [3] or the insulating transition in quasi-disordered optical lattice [4, 5] have been studied in this manner. More recently, transport in quantum gases is also studied in junction-type setup more analogous to condensed-matter systems: two reservoirs with different chemical potentials are connected through a constriction. For example, in a gas of fermions, the quantization of conductance through a quantum point contact [6] and the superfluid to normal transition in a disordered thin film have been observed [7]. In our work, we focus on the transport of non-interacting particles in disordered media. Without a bias force, quantum interferences between multiple paths lead to Anderson localization [8] whose signature is an exponential decay in space of single particle wave-function [9]. This phenomenon is ubiquitous in wave/quantum physics and it has been observed in many physical contexts [10] including condensed-matter [11] and ultra-cold atoms [12-14]. One-dimensional truly disordered systems are always localized [15], contrary to the 3D case where a phase transition between localized and extended single particle wave-functions takes place as a function of the disorder strength [16-18]. The localization properties of 1D disordered systems are modified in the presence of a bias force. Theoretical studies predict a transition from algebraic localization to delocalization as a function of a single control non-dimensional parameter $\alpha$ which is the ratio of the force to the disorder strength [19, 20]. Physically, $\alpha$ is the relative energy gain $\Delta$E/E of a particle of energy E when moving over a localization length. Interestingly, in a 1D white noise disorder, this quantity is independent of E as the localization length is proportional to E. If $\alpha$ is small, the force does not considerably change the localization behavior of the particle while for large $\alpha$ its dynamics is severely affected leading to delocalization. This localization-delocalization transition is predicted in the infinite time limit for white noise disorder [20]. In a correlated disorder, as the one produced from a far-off-resonance laser speckle [21], the situation is more complicated. Speckles have no Fourier component beyond a spatial frequency 2k c. As a consequence, back-scattering and localization are not expected in the framework of Born approximation for atoms with wavevectors k > k c [12, 22]. Since localized wave-functions always have a small fraction at long distance corresponding to large energies and momenta in the presence of a bias force, we thus expect correlation-induced delocalization at infinite time. However, signatures of the algebraic localization-delocalization transition are predicted to be observable at transient times [20]. In this paper, we report on the observation of the algebraic localization-delocalization transition with cold-atoms propagating in a one dimensional disordered potential in the presence of a controlled bias force. We experimentally show that the non-dimensional parameter $\alpha$ is the only relevant parameter to describe the transition. We notice that the initial velocity of the quantum wave packet only plays a role through the correlation of the disordered potential, showing that the transition is in-trinsically energy independent. In the localized regime, we demonstrate an algebraic decay of the density and measure the corresponding decay exponent as a function of $\alpha$. At large disorder strength, a saturation of the expo
Figures
Figures from the paper (1 more)
Reference graph
Works this paper leans on
- [1]
-
[2]
M. Ben Dahan, E. Peik, J. Reichel, Y. Castin, and C. Salomon, Phys. Rev. Lett., 76, 4508 (1996)
work page 1996
- [3]
- [4]
-
[5]
C. D’Errico, E. Lucioni, L. Tanzi, L. Gori, G. Roux, I.P. McCulloch, T. Giamarchi, M. Inguscio, and G. Modugno, Phys. Rev. Lett. 113, 095301 (2014)
work page 2014
-
[6]
S. Krinner, D. Stadler, D. Husmann, J.-P. Brantut, T. Esslinger, Nature 517, 64 (2015)
work page 2015
-
[7]
S. Krinner, D. Stadler, J. Meineke, J.-P. Brantut, and T. Esslinger, Phys. Rev. Lett. 110, 100601 (2013)
work page 2013
- [8]
Show all 32 references
-
[9]
Sanchez-Palencia, D
L. Sanchez-Palencia, D. Cl´ ement, P. Lugan, P. Bouyer, G. V. Shlyapnikov, and A. Aspect, Phys. Rev. Lett. 98, 210401 (2007)
2007
-
[10]
Lagendijk, B
A. Lagendijk, B. van Tiggelen, D. Wiersma, Physics To- day 62, 24 (2009)
2009
-
[11]
Gomez-Navarro, P.J
C. Gomez-Navarro, P.J. De Pablo, J. Gomez-Herrero, B. Biel, F.J. Garcia-Vidal, A. Rubio and F. Flores, Nature Mat. 4, 534 (2005)
2005
-
[12]
Billy, V
J. Billy, V. Josse, Z. Zuo, A. Bernard, B. Hambrecht, P. Lugan, D. Cl´ ement, L. Sanchez-Palencia, P. Bouyer, and A. Aspect, Nature 453, 891 (2008)
2008
-
[13]
Chab´ e, G
J. Chab´ e, G. Lemari´ e, B. Gr´ emaud, D. Delande, P. Szriftgiser, J.-C. Garreau, Phys. Rev. Lett. 101, 255702 (2008)
2008
-
[14]
Roati, C
G. Roati, C. D’Errico, L. Fallani, M. Fattori, C. Fort, M. Zaccanti, G. Modugno, M. Modugno, and M. Inguscio, Nature 453, 895 (2008)
2008
-
[15]
Abrahams, P.W
E. Abrahams, P.W. Anderson, D.C. Licciardello, and T.V. Ramakrishnan, Phys. Rev. Lett. 42, 673 (1979)
1979
-
[16]
Lopez, J.-F
M. Lopez, J.-F. Cl´ ement, P. Szriftgiser, J. C. Garreau and D. Delande, Phys. Rev. Lett. 108, 095701 (2012)
2012
-
[17]
Jendrzejewski, A
F. Jendrzejewski, A. Bernard, K. M¨ uller, P. Cheinet, V. Josse, M. Piraud, L. Pezze, L. Sanchez-Palencia, A. As- pect and P. Bouyer, Nature Physics 8, 392 (2012)
2012
-
[18]
Kondov, W.R
S.S. Kondov, W.R. McGehee, J.J. Zirbel, B. DeMarco, Science 334, 66 (2011)
2011
-
[19]
Crosnier de Bellaistre, A
C. Crosnier de Bellaistre, A. Aspect, A. Georges, and L. Sanchez-Palencia E, Phys. Rev. B, 95, 140201(R) (2017)
2017
-
[20]
Crosnier de Bellaistre, C
C. Crosnier de Bellaistre, C. Trefzger, A. Aspect, A. Georges, and L. Sanchez-Palencia. Phys. Rev. A 97, 013613 (2018)
2018
-
[21]
Cl´ ement, A.F
D. Cl´ ement, A.F. Var´ on, J.A. Retter, L. Sanchez- Palencia, A. Aspect and P. Bouyer, New J. Phys. 8, 165 (2006)
2006
-
[22]
Lugan, A
P. Lugan, A. Aspect, L. Sanchez-Palencia, D. Delande, B. Gr´ emaud, C. A. M¨ uller, and C. Miniatura, Phys. Rev. A 80, 023605 (2009)
2009
-
[23]
D’Errico, M
C. D’Errico, M. Zaccanti, M. Fattori, G. Roati, M. In- guscio, G. Modugno, A. Simoni, New J. of Phys. 9, 223 (2007)
2007
-
[24]
Landini, S
M. Landini, S. Roy, G. Roati, A. Simoni, M. Inguscio, G. Modugno, and M. Fattori, Phys. Rev. A 86, 033421 (2012)
2012
-
[25]
Salomon, L
G. Salomon, L. Fouch´ e, S. Lepoutre, A. Aspect, and T. Bourdel, Phys. Rev. A 90, 033405 (2014)
2014
-
[26]
Lepoutre, L
S. Lepoutre, L. Fouch´ e, A. Boiss´ e, G. Berthet, G. Sa- lomon, A. Aspect, and T. Bourdel, Phys. Rev. A 94, 053626 (2016)
2016
-
[27]
Boiss´ e, G
A. Boiss´ e, G. Berthet, L. Fouch, G. Salomon, A. Aspect, S. Lepoutre, T. Bourdel, Euro. Phys. Lett., 117 10007 (2017)
2017
-
[28]
We thus limit the propagation time of our experiment in order that the atoms never reach the necessary velocity of 9.6 mm.s−1 corresponding to half of the lattice wavevec- tor
Due to its normal incidence to the science chamber and its large coherence length, the 1064 nm laser beam pro- duces a sub-nanoKelvin residual periodic lattice on the atoms that can lead to Bloch oscillations [2]. We thus limit the propagation time of our experiment in order t...
-
[29]
T. Prat, N. Cherroret, and D. Delande, Phys. Rev. A 94, 022114 (2016)
2016
-
[30]
Volchkov, M
V.V. Volchkov, M. Pasek, V. Denechaud, M. Mukhtar, A. Aspect, D. Delande and V. Josse, Phys. Rev. Lett. 120, 060404 (2018)
2018
-
[31]
Aleiner, B.L
I.L. Aleiner, B.L. Altshuler, and G.V. Shlyapnikov, Na- ture Physics 6, 900 (2010)
2010
-
[32]
Giamarchi and H
T. Giamarchi and H. J. Schulz, Phys. Rev. B 37, 325 (1988)
1988
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.