Pith. sign in

REVIEW 3 major objections 6 minor 2 cited by

Repulsive barricade pushes tweezer loading past 80%

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 · glm-5.2

2026-07-04 17:43 UTC pith:OOOYZ7MJ

load-bearing objection dark-state shelving approaches in Refs. 39 and 41. It's species-agnostic, which is the real selling point: molecules like CaF are stuck at ~35% fill fractions and don't have obvious shelving states, so this fills a genuine gap. the 3 major comments →

arxiv 2604.22406 v2 pith:OOOYZ7MJ submitted 2026-04-24 physics.atom-ph physics.comp-ph

Near-deterministic loading of optical tweezer arrays via repulsive barricade potentials

Archie C. Baldock , Alex J. Matthies , Luke Caldwell , Hannah J. Williams This is my paper
classification physics.atom-ph physics.comp-ph PACS 37.10.Gh37.10.Vz32.80.Wr
keywords loadingarraysatomscyclesenablingfillingmoleculesoptical
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 proposes a scheme for breaking through the stochastic loading limit of optical tweezer arrays — the ceiling of roughly 50 percent for atoms and 35 percent for molecules that arises because light-assisted collisions eject pairs from each trap. The central object is the barricade tweezer: a trap formed by superimposing a standard attractive (red-detuned) optical tweezer with a repulsive (blue-detuned) beam of larger waist, creating a potential landscape with a trapping core surrounded by a repulsive shell. Once a particle is loaded and identified, the barricade is ramped on, preventing new particles from entering the core during subsequent cooling and loading cycles. The paper uses three-dimensional semi-classical Monte Carlo trajectory simulations, parameterized for 87Rb atoms and CaF molecules, to show that the barricade suppresses the flux of incoming particles by several orders of magnitude, yielding collision-limited lifetimes of approximately 1.3 seconds for both species. Plugging these lifetimes into a cumulative loading model, the authors predict that four loading cycles would raise the filling fraction to 87 percent for Rb and 81 percent for CaF, up from single-cycle values of 50 and 35 percent. The key practical claim is that the required optical powers — a few milliwatts for each beam — are comparable to standard tweezer powers, and that the filling fraction saturates rapidly with increasing barrier height, meaning the theoretical maxima are accessible without exotic hardware. The scheme is species-independent, requiring only that attractive and repulsive optical potentials can be simultaneously engineered.

Core claim

The paper's central result is that a repulsive optical barricade, formed by overlapping a blue-detuned beam with a standard red-detuned tweezer, can suppress the flux of particles entering an already-occupied trap by over three orders of magnitude. Using 3D Monte Carlo simulations calibrated to experimental parameters for 87Rb and CaF, the authors show that this suppression corresponds to collision-limited lifetimes of about 1.3 seconds in both cases. With cycle times of 75 ms (Rb) and 15 ms (CaF), four cumulative loading cycles would yield filling fractions of 87 percent and 81 percent respectively, approaching the theoretical maxima of 94 percent and 82 percent for four cycles in the limit

What carries the argument

The barricade tweezer potential is the sum of an attractive light shift (red-detuned beam, diffraction-limited waist) and a repulsive light shift (blue-detuned beam, larger waist). The repulsive beam creates a shell-like barrier around the trapping core. Flux suppression is modeled by 3D semi-classical Monte Carlo trajectory simulations of particles drawn from a Maxwell-Boltzmann distribution evolving in the combined potential. An analytical isotropic barrier model (complementary error function plus exponential term) is shown to fit the Monte Carlo results well once the barrier is sufficiently isotropic. The cumulative fill fraction follows a linear recurrence relation accounting for three f

Load-bearing premise

The simulation assumes the repulsive barricade can be ramped on adiabatically without heating or losing the already-trapped particle. If intensity noise, SLM switching artifacts, or non-adiabatic trap-frequency changes during the ramp cause significant heating, the protected particle would be lost before the next loading cycle, and the cumulative fill fraction model would overestimate performance.

What would settle it

Experimentally, if the measured collision-limited lifetime of a protected particle is substantially shorter than 1.3 s — for instance, because the ramp-on of the repulsive beam causes parametric heating or because off-resonant scattering from the blue-detuned light is higher than estimated — then the cumulative fill fractions would fall below the predicted 87 percent and 81 percent, and the advantage of the barricade scheme over existing dark-state protection methods would diminish.

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

If this is right

  • If the barricade scheme works experimentally as simulated, molecule tweezer arrays — currently limited to 35 percent filling — could reach over 80 percent filling, removing a major bottleneck for molecular quantum simulation and computation.
  • The scheme is species-independent and requires no internal-state shelving, so it could be applied to any atom or molecule for which both attractive and repulsive optical potentials can be engineered at compatible wavelengths.
  • Combined with existing parallel rearrangement techniques, near-unity filling of large defect-free arrays becomes feasible, directly enabling scaling to larger quantum technology platforms.
  • The collision-limited lifetime of 1.3 s sets a ceiling on the number of useful loading cycles; extending this lifetime (via higher barriers or lower-density reservoirs) would push filling fractions closer to unity.

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 / 6 minor

Summary. This manuscript proposes a 'barricade tweezer' scheme for near-deterministic loading of optical tweezer arrays via multiple loading cycles. The core idea is to superimpose a blue-detuned repulsive beam with a red-detuned attractive tweezer to create a potential barrier around already-loaded sites, preventing new particles from entering and causing light-assisted collisions. The authors perform 3D semi-classical Monte Carlo trajectory simulations to compute the flux suppression through the trapping core for 87Rb and CaF, and use an analytical isotropic barrier model (Eq. 8) to fit the results. They report collision-limited lifetimes of ~1.3 s for both species, yielding cumulative fill fractions of 87% (Rb) and 81% (CaF) after four loading cycles. The fill fraction model (Eq. 1) is derived from a recursion relation (Eq. 4) in the Appendix. The scheme is general and does not require long-lived shelving states, distinguishing it from prior dark-state protection proposals.

Significance. The paper addresses a practical bottleneck in tweezer array scaling — the stochastic loading limit — with a conceptually simple and broadly applicable protocol. The Monte Carlo simulations use realistic experimental parameters and the code is publicly available (GitHub), which is a notable strength for reproducibility. The conservative choices for CaF (R_core = r_FWHM, worst-case m_F state) are appropriate and lend credibility to the predictions. The isotropic barrier model (Eq. 8) provides a useful analytical framework. The proposal is timely given the rapid scaling of tweezer arrays to thousands of sites and the difficulty of extending collisional shielding to molecules.

major comments (3)
  1. The central claim that cumulative fill fractions exceed 80% after four cycles depends on the barricade being ramped on without loss or heating of the already-trapped particle. The manuscript states this will be done 'adiabatically' (Fig. 1b, main text) but provides no analysis of the ramp dynamics: no estimate of the required ramp time, no simulation of the time-dependent potential during ramping, and no discussion of whether the trap frequencies change discontinuously as the repulsive beam is introduced. For typical tweezer parameters this is likely feasible, but the simultaneous power increase of the attractive beam to maintain trap depth creates a time-dependent potential landscape that could excite or eject the trapped particle. A brief analysis — even an order-of-magnitude estimate of the adiabaticity condition (ramp time vs. inverse trap frequency) and a discussion of potential SLM
  2. switching artifacts or intensity noise — would substantially strengthen the paper, as this is the single most critical unvalidated assumption underpinning the cumulative fill fraction model.
  3. Section 'CaF', paragraph 4: The manuscript notes that 'MHz-rate photon scattering during imaging and cooling causes a molecule to traverse the full m_F manifold many times while crossing the barrier,' justifying the use of the worst-case m_F state. However, the time scale for m_F mixing relative to the barrier transit time is not quantified. If the molecule scatters photons at MHz rates but the barrier transit time is sub-microsecond, the molecule may not fully explore the m_F manifold during a single crossing attempt. A brief estimate of the ratio of scattering rate to transit rate would make this conservative assumption more convincing.
minor comments (6)
  1. Eq. (1): The notation η(τ) is used both for the reload probability and appears in the same equation as η (the cumulative fill fraction). Consider renaming the reload probability (e.g., to p_reload) to avoid confusion.
  2. Appendix, Eq. (5): The statement 'which recovers the saturation limit of η_0 = 0.5 in the limit of t/τ ≫ 1' should be clarified — Eq. (5) gives η(τ) → 1/2 in that limit, but η_0 is defined separately as the single-cycle loading probability. The connection should be stated more precisely.
  3. Fig. 3: The axis labels in panels (c) and (d) are difficult to read in the rendered version. Consider increasing font sizes or simplifying the layout.
  4. Section 'CaF': The footnote defining the F = 1⁻ state (footnote 3) uses a superscript that could be confused with a reference. Consider using a different notation (e.g., F = 1_lower).
  5. The manuscript uses both 'filling fraction' and 'fill fraction' interchangeably. Standardizing on one term would improve consistency.
  6. Reference [51] (Humphreys et al., CPC 2025) appears to be by co-authors of this manuscript; this relationship should be disclosed if the journal requires it.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading and positive assessment. The referee raises two substantive points: (1) the need for an analysis of the adiabatic ramp-on dynamics of the barricade potential, and (2) a quantitative justification for the assumption that molecules explore the full m_F manifold during barrier transit. Both points are well-taken, and we will address them in a revised manuscript by adding an order-of-magnitude adiabaticity estimate and a scattering-rate-to-transit-rate comparison for CaF.

read point-by-point responses

  1. Referee: Major Comment 1: The central claim depends on the barricade being ramped on without loss or heating. No analysis of ramp dynamics, adiabaticity condition, trap frequency changes, or SLM switching artifacts is provided.

    Authors: The referee is correct that the ramp-on dynamics of the barricade potential are not analyzed in the current manuscript, and that this is a critical assumption underpinning the cumulative fill fraction model. We agree that an explicit discussion is needed. In the revised manuscript, we will add an order-of-magnitude adiabaticity analysis. The key consideration is that the trapped particle sits in the ground state of the attractive tweezer (radial trap frequency ~100 kHz for both Rb and CaF with the parameters used), so the adiabaticity condition requires ramp times much longer than ~10 microseconds. SLM switching times are typically ~1 ms or faster with modern devices (as cited in Ref. 42), and intensity ramps over ~1-10 ms are standard experimental practice, comfortably satisfying the adiabaticity condition. We will also note that the trap depth is maintained constant during the ramp (as described in the manuscript), so the trap frequency changes are modest rather than discontinuous. We will add a brief discussion of potential parametric heating from intensity noise on the repulsive beam, noting that the heating rate from off-resonant scattering (already estimated at 3-4 microK/s) dominates over any parametric heating contribution at these ramp times. This will be added as a short subsection or paragraph in the Methods section. revision: partial

  2. Referee: Major Comment 2: The m_F mixing argument for CaF is not quantified. If barrier transit time is sub-microsecond but scattering is MHz-rate, the molecule may not fully explore the m_F manifold during a single crossing attempt.

    Authors: This is a fair point. We will add a quantitative estimate. For CaF, the photon scattering rate during imaging/cooling is on the order of 1 MHz (consistent with the manuscript's statement). The barrier transit time can be estimated as the barrier width divided by the thermal velocity: for a barrier width of ~1 micron and a thermal velocity of ~0.3 m/s (at 9 microK), the transit time is ~3 microseconds. This gives roughly 3 photon scattering events per transit, which is indeed marginal for full m_F manifold exploration during a single crossing. However, the relevant comparison is not for a single crossing attempt but over the full loading cycle (10 ms of cooling), during which a molecule near the barrier scatters ~10,000 photons and traverses the m_F manifold many times. The worst-case m_F state assumption is therefore conservative in the sense that it represents a lower bound on barrier effectiveness for any individual encounter, and over many encounters the molecule samples all m_F states. We will add this estimate to the CaF section to make the conservatism of the assumption explicit, and we will soften the claim from 'traverse the full m_F manifold many times while crossing the barrier' to a more precise statement about the number of scattering events per transit and the cumulative exploration over the loading cycle. revision: partial

Circularity Check

0 steps flagged

No significant circularity found; derivation chain is self-contained

full rationale

The paper's derivation chain proceeds in two independent stages: (1) a first-principles optical potential (Eqs. 2-3, vectorial non-paraxial model) feeds a 3D Monte Carlo trajectory simulation that extracts particle flux through the trapping core, from which the collision-limited lifetime τ = 1/flux is obtained; (2) τ is fed into the fill fraction model (Eq. 1), which is independently derived from a linear recurrence relation (Eq. 4) in the Appendix. No step defines its output in terms of itself. The isotropic barrier model (Eq. 8) is fitted to the Monte Carlo data with j₀ and T as free parameters, but it is used only as an analytical interpolation of the simulation results (shown as dashed lines in Fig. 3), not presented as an independent prediction. The fill fraction recursion (Eq. 4) is solved correctly as a linear recurrence with constant coefficients, and the η(τ) term (Eqs. 5-6) is negligible at the operating parameters (t/τ ≈ 0.06 for Rb, 0.01 for CaF), so even its heuristic treatment for molecules is not load-bearing. The only self-citation is Ref [51] (Humphreys, Matthies, Williams — two co-authors of this paper), cited for tensor polarizability computation methods for molecules; this is a methods/tooling citation, not a load-bearing theoretical premise. The central results (τ ≈ 1.3 s, η > 80%) follow from the Monte Carlo simulation and the independently derived fill fraction formula, with no reduction to inputs by construction.

Axiom & Free-Parameter Ledger

3 free parameters · 3 axioms · 0 invented entities

The ledger captures the fitted parameters in the isotropic barrier model and the key domain assumptions. No new physical entities are invented; the scheme uses established optical components and known atomic/molecular species.

free parameters (3)
  • j0 (unperturbed flux)
    Free parameter in the isotropic barrier model fit (Eq. 8), representing the flux in the absence of a barrier.
  • T (effective temperature)
    Free parameter in the isotropic barrier model fit (Eq. 8), representing the effective temperature of the reservoir.
  • R_core (trapping core radius) = 50 nm (Rb), 0.63 µm (CaF)
    Chosen by hand to represent the Condon radius for Rb and r_FWHM for CaF; affects the calculated flux.
axioms (3)
  • domain assumption Light-assisted collisions cause pairwise ejection, saturating loading at 50% for atoms.
    Standard assumption from Ref 31, used to derive the fill fraction recursion (Eq. 4) and the reload probability (Eq. 5).
  • domain assumption The barricade potential can be ramped on adiabatically without parametric heating.
    Invoked in the protection sequence description (Fig 1b, Methods). This is a load-bearing premise for the feasibility of the scheme.
  • ad hoc to paper The same pairwise ejection mechanism operates for molecules, with an additional unknown loss mechanism.
    Used to justify the heuristic generalization of the reload probability for molecules (Eq. 6), which is needed to match the observed 35% saturation.

pith-pipeline@v1.1.0-glm · 14125 in / 2300 out tokens · 183602 ms · 2026-07-04T17:43:11.443570+00:00 · methodology

0 comments
read the original abstract

Optical tweezers are a powerful tool for creating defect-free arrays of atoms and molecules, enabling advances in quantum simulation, computation, and precision metrology. However, the achievable array size is limited by the initial loading fraction, typically $50\,\%$ for atoms and $35\,\%$ for molecules. Here, we propose a general scheme for enabling multiple loading cycles by protecting trapped particles using a repulsive barrier. We show that collision-limited lifetimes of particles in protected tweezers can exceed one second, leading to filling fractions of over $80\%$ after four loading cycles. Combined with existing rearrangement techniques, this approach enables efficient unity filling of tweezer arrays and provides a scalable pathway towards larger quantum technology platforms.

Figures

Figures reproduced from arXiv: 2604.22406 by Alex J. Matthies, Archie C. Baldock, Hannah J. Williams, Luke Caldwell.

Figure 1
Figure 1. Figure 1: FIG. 1. (a) Illustration of the barricade tweezer set up. Spatial light modulators (SLMs) are used to control the phase view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. (a) Optical potential, view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. (a,b) Simulated particle flux through the core of the barricade tweezer as a function of radial barrier height for (a) view at source ↗

Review history (2 revisions) →

discussion (0)

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

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Deterministic loading of molecular arrays by microwave-assisted collisions

    physics.atom-ph 2026-07 conditional novelty 7.0

    Microwave-assisted collisions with repulsive van der Waals shelving are predicted to load single molecules into tweezers at 87-96% filling fractions, up from the current 30-40%.

  2. Deterministic loading of molecular arrays by microwave-assisted collisions

    physics.atom-ph 2026-07 conditional novelty 7.0

    Shelving molecules in rotationally or vibrationally excited states and driving microwave-assisted collisions can suppress loss, eject one molecule deterministically, and yield predicted filling fractions up to 96%.

Reference graph

Works this paper leans on

51 extracted references · 51 canonical work pages · cited by 1 Pith paper

  1. [1]

    Browaeys and T

    A. Browaeys and T. Lahaye, Nat. Phys.16, 132 (2020)

    work page 2020

  2. [2]

    A. M. Kaufman and K.-K. Ni, Nature Physics17, 1324 (2021)

    work page 2021

  3. [3]

    Endres, H

    M. Endres, H. Bernien, A. Keesling, H. Levine, E. R. Anschuetz, A. Krajenbrink, C. Senko, V. Vuletic, M. Greiner, and M. D. Lukin, Science354, 1024 (2016)

    work page 2016

  4. [4]

    Barredo, S

    D. Barredo, S. de L´ es´ eleuc, V. Lienhard, T. Lahaye, and A. Browaeys, Science354, 1021 (2016)

    work page 2016

  5. [5]

    W. Lee, H. Kim, and J. Ahn, Opt. Express24, 9816 (2016)

    work page 2016

  6. [6]

    M. A. Norcia, A. W. Young, and A. M. Kaufman, Phys. Rev. X8, 041054 (2018)

    work page 2018

  7. [7]

    Saskin, J

    S. Saskin, J. T. Wilson, B. Grinkemeyer, and J. D. Thompson, Phys. Rev. Lett.122, 143002 (2019)

    work page 2019

  8. [8]

    N. C. Jackson, R. K. Hanley, M. Hill, F. Leroux, C. S. Adams, and M. P. A. Jones, SciPost Phys.8, 038 (2020)

    work page 2020

  9. [9]

    Sheng, J

    C. Sheng, J. Hou, X. He, K. Wang, R. Guo, J. Zhuang, B. Mamat, P. Xu, M. Liu, J. Wang, and M. Zhan, Phys. Rev. Lett.128, 083202 (2022)

    work page 2022

  10. [10]

    Anderegg, L

    L. Anderegg, L. W. Cheuk, Y. Bao, S. Burchesky, W. Ketterle, K.-K. Ni, and J. M. Doyle, Science365, 1156 (2019)

    work page 2019

  11. [11]

    Y. Bao, S. S. Yu, L. Anderegg, E. Chae, W. Ketterle, K.-K. Ni, and J. M. Doyle, Science382, 1138 (2023)

    work page 2023

  12. [12]

    C. M. Holland, Y. Lu, and L. W. Cheuk, Phys. Rev. Lett. 131, 053202 (2023)

    work page 2023

  13. [13]

    A. J. Park, L. R. B. Picard, G. E. Patenotte, J. T. Zhang, T. Rosenband, and K.-K. Ni, Phys. Rev. Lett. 131, 183401 (2023)

    work page 2023

  14. [14]

    N. B. Vilas, P. Robichaud, C. Hallas, G. K. Li, L. An- deregg, and J. M. Doyle, Nature628, 282 (2024)

    work page 2024

  15. [15]

    D. K. Ruttley, T. R. Hepworth, A. Guttridge, and S. L. Cornish, Nature637, 827 (2025)

    work page 2025

  16. [16]

    Nogrette, H

    F. Nogrette, H. Labuhn, S. Ravets, D. Barredo, L. B´ eguin, A. Vernier, T. Lahaye, and A. Browaeys, Phys. Rev. X4, 021034 (2014). 6

    work page 2014

  17. [17]

    H. Kim, W. Lee, H. G. Lee, H. Jo, Y. Song, and J. Ahn, Nature Communications7, 13317 (2016)

    work page 2016

  18. [18]

    Schymik, V

    K.-N. Schymik, V. Lienhard, D. Barredo, P. Scholl, H. Williams, A. Browaeys, and T. Lahaye, Phys. Rev. A102, 063107 (2020)

    work page 2020

  19. [19]

    Bluvstein, H

    D. Bluvstein, H. Levine, G. Semeghini, T. T. Wang, S. Ebadi, M. Kalinowski, A. Keesling, N. Maskara, H. Pichler, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Nature604, 451 (2022)

    work page 2022

  20. [20]

    C. M. Holland, Y. Lu, and L. W. Cheuk, Science382, 1143 (2023)

    work page 2023

  21. [21]

    H. J. Manetsch, G. Nomura, E. Bataille, X. Lv, K. H. Leung, and M. Endres, Nature647, 60 (2025)

    work page 2025

  22. [22]

    Pichard, D

    G. Pichard, D. Lim, E. Bloch, J. Vaneecloo, L. Boura- chot, G.-J. Both, G. M´ eriaux, S. Dutartre, R. Hostein, J. Paris, B. Ximenez, A. Signoles, A. Browaeys, T. La- haye, and D. Dreon, Phys. Rev. Appl.22, 024073 (2024)

    work page 2024

  23. [23]

    N.-C. Chiu, E. C. Trapp, J. Guo, M. H. Abobeih, L. M. Stewart, S. Hollerith, P. L. Stroganov, M. Kalinowski, A. A. Geim, S. J. Evered, S. H. Li, X. Lyu, L. M. Peters, D. Bluvstein, T. T. Wang, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Nature646, 1075 (2025)

    work page 2025

  24. [24]

    Ebadi, T

    S. Ebadi, T. T. Wang, H. Levine, A. Keesling, G. Se- meghini, A. Omran, D. Bluvstein, R. Samajdar, H. Pich- ler, W. W. Ho, S. Choi, S. Sachdev, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Nature595, 227 (2021)

    work page 2021

  25. [25]

    Scholl, M

    P. Scholl, M. Schuler, H. J. Williams, A. A. Eberharter, D. Barredo, K.-N. Schymik, V. Lienhard, L.-P. Henry, T. C. Lang, A. M. L¨ auchli, T. Lahaye, and A. Browaeys, Nature595, 233 (2021)

    work page 2021

  26. [26]

    T. M. Graham, Y. Song, J. Scott, C. Poole, L. Phut- titarn, P. Jooya, K.and Eichler, X. Jiang, A. Marra, B. Grinkemeyer, M. Kwon, M. Ebert, J. Cherek, M. T. Lichtman, M. Gillette, J. Gilbert, D. Bowman, T. Bal- lance, C. Campbell, E. D. Dahl, O. Crawford, N. S. Blunt, B. Rogers, T. Noel, and M. Saffman, Nature604, 457 (2022)

    work page 2022

  27. [27]

    S. J. Evered, D. Bluvstein, M. Kalinowski, S. Ebadi, T. Manovitz, H. Zhou, S. H. Li, A. A. Geim, T. T. Wang, N. Maskara, H. Levine, G. Semeghini, M. Greiner, V. Vuleti´ c, and M. D. Lukin, Nature622, 268 (2023)

    work page 2023

  28. [28]

    Bluvstein, A

    D. Bluvstein, A. A. Geim, S. H. Li, S. J. Evered, J. P. B. Ataides, G. Baranes, A. Gu, T. Manovitz, M. Xu, M. Kalinowski, S. Majidy, C. Kokail, N. Maskara, E. C. Trapp, L. M. Stewart, S. Hollerith, H. Zhou, M. J. Gul- lans, S. F. Yelin, M. Greiner, V. Vuleti´ c, M. Cain, and M. D. Lukin, Nature649, 39 (2026)

    work page 2026

  29. [29]

    I. S. Madjarov, A. Cooper, A. L. Shaw, J. P. Covey, V. Schkolnik, T. H. Yoon, J. R. Williams, and M. En- dres, Phys. Rev. X9, 041052 (2019)

    work page 2019

  30. [30]

    A. W. Young, W. J. Eckner, W. R. Milner, D. Kedar, M. A. Norcia, E. Oelker, N. Schine, J. Ye, and A. M. Kaufman, Nature588, 408 (2020)

    work page 2020

  31. [31]

    Schlosser, G

    N. Schlosser, G. Reymond, and P. Grangier, Phys. Rev. Lett.89, 023005 (2002)

    work page 2002

  32. [32]

    Schlosser, G

    N. Schlosser, G. Reymond, I. Protsenko, and P. Grangier, Nature411, 1024 (2001)

    work page 2001

  33. [33]

    Gr¨ unzweig, A

    T. Gr¨ unzweig, A. Hilliard, M. McGovern, and M. F. An- dersen, Nat. Phys.6, 951 (2010)

    work page 2010

  34. [34]

    B. J. Lester, N. Luick, A. M. Kaufman, C. M. Reynolds, and C. A. Regal, Phys. Rev. Lett.115, 073003 (2015)

    work page 2015

  35. [35]

    M. M. Aliyu, L. Zhao, X. Q. Quek, K. C. Yellapragada, and H. Loh, Phys. Rev. Res.3, 043059 (2021)

    work page 2021

  36. [36]

    M. O. Brown, T. Thiele, C. Kiehl, T.-W. Hsu, and C. A. Regal, Phys. Rev. X9, 011057 (2019)

    work page 2019

  37. [37]

    Ang’ong’a, C

    J. Ang’ong’a, C. Huang, J. P. Covey, and B. Gadway, Phys. Rev. Res.4, 013240 (2022)

    work page 2022

  38. [38]

    Jenkins, J

    A. Jenkins, J. W. Lis, A. Senoo, W. F. McGrew, and A. M. Kaufman, Phys. Rev. X12, 021027 (2022)

    work page 2022

  39. [39]

    E. F. Walraven, M. R. Tarbutt, and T. Karman, Phys. Rev. Lett.132, 183401 (2024)

    work page 2024

  40. [40]

    I. H. A. Knottnerus, Y. C. Tseng, A. Urech, R. J. C. Spreeuw, and F. Schreck, SciPost Phys.19, 118 (2025)

    work page 2025

  41. [41]

    A. L. Shaw, P. Scholl, R. Finklestein, I. S. Madjarov, B. Grinkemeyer, and M. Endres, Phys. Rev. Lett.130, 193402 (2023)

    work page 2023

  42. [42]

    Vanackere, A

    T. Vanackere, A. Hermans, I. Christen, C. Panuski, M. Dong, M. Zimmermann, H. Raniwala, A. J. Leen- heer, M. Eichenfield, G. Gilbert, and D. Englund, Nature Communications16, 11637 (2025)

    work page 2025

  43. [43]

    C. S. Adams and I. G. Hughes,Optics f2f: From Fourier to Fresnel(Oxford University Press, 2018)

    work page 2018

  44. [44]

    Y. H. Fung, P. Sompet, and M. F. Andersen, Technologies 4, 4 (2016)

    work page 2016

  45. [45]

    Fuhrmanek, R

    A. Fuhrmanek, R. Bourgain, Y. R. P. Sortais, and A. Browaeys, Phys. Rev. A85, 062708 (2012)

    work page 2012

  46. [46]

    Grimm, M

    R. Grimm, M. Weidem¨ uller, and Y. B. Ovchinnikov, Ad- vances in Atomic, Molecular, and Optical Physics42, 95 (2000)

    work page 2000

  47. [47]

    E. B. Norrgard, D. J. McCarron, M. H. Steinecker, M. R. Tarbutt, and D. DeMille, Phys. Rev. Lett.116, 063004 (2016)

    work page 2016

  48. [48]

    H. J. Williams, S. Truppe, M. Hambach, L. Caldwell, N. J. Fitch, E. A. Hinds, B. E. Sauer, and M. R. Tarbutt, New J. of Phys.19, 113035 (2017)

    work page 2017

  49. [49]

    Anderegg, B

    L. Anderegg, B. L. Augenbraun, E. Chae, B. Hemmer- ling, N. R. Hutzler, A. Ravi, A. Collopy, J. Ye, W. Ket- terle, and J. M. Doyle, Phys. Rev. Lett.119, 103201 (2017)

    work page 2017

  50. [50]

    A. L. Collopy, S. Ding, Y. Wu, I. A. Finneran, L. An- deregg, B. L. Augenbraun, J. M. Doyle, and J. Ye, Phys. Rev. Lett.121, 213201 (2018)

    work page 2018

  51. [51]

    Humphreys, A

    B. Humphreys, A. J. Matthies, and H. J. Williams, Com- puter Physics Communications316, 109813 (2025). Appendix: Fill F raction Model We derive the cumulative fill fractionηafterNload- ing cycles by considering the probabilityP N that a given tweezer site is occupied at the end of theN-th loading cycle. In the first cycle, no barricade is present and thus...

    work page 2025