Enhancing the capture velocity of a Dy magneto-optical trap with two-stage slowing

Alan O. Jamison; Li Du; Michael Cantara; Pierre Barral; William Lunden; Wolfgang Ketterle

REVIEW 2 major objections 5 minor 19 references

Enhancing the capture velocity of a Dy magneto-optical trap with two-stage slowing

T0 review · 2 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read A pair of low-power beams crossing just in front of the MOT lets a narrow-line dysprosium trap load 3×10^8 atoms in 2 seconds, a more than 20-fold improvement over Zeeman slowing alone.

desk verdict Useful Dy MOT loading technique, but the headline >20x gain is not cleanly isolated from transverse cooling in the data as presented. read the letter →

arxiv 1908.10433 v1 pith:26EGL75F submitted 2019-08-27 cond-mat.quant-gas physics.atom-ph

William Lunden , Li Du , Michael Cantara , Pierre Barral , Alan O. Jamison , Wolfgang Ketterle This is my paper · ORCID

classification cond-mat.quant-gasphysics.atom-ph PACS 32.80.Pj37.10.De37.10.Gh

keywords magneto-opticaltrapnarrow-linecoolingZeemanslowerangledslowingdysprosiumMOTloadingratetransversebeamspreadinglaser

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that a two-stage 'angled slowing' scheme can overcome the main bottleneck in loading a narrow-line dysprosium magneto-optical trap: the tendency of a Zeeman-slowed atomic beam to spread transversely and miss the trap during free flight. The authors show that slowing the beam with two low-power beams crossing just in front of the MOT, instead of relying on the Zeeman slower alone, raises the MOT population by a factor greater than 20 and loads about 3×$10^{8}$ atoms in 2 seconds. This matters because the narrow 626 nm Dy transition has a low Doppler temperature but also a low capture velocity (roughly 8 m/s), so simple Zeeman slowing loses most of the slowed flux before it reaches the MOT; the technique offers a low-power way around this that should transfer to other narrow-line species.

What carries the argument

The mechanism is the 'angled slowing' beam pair: two near-resonant 421 nm beams, angled symmetrically about the atomic beam axis so their transverse radiation-pressure components cancel, that apply a net longitudinal slowing force to atoms a few centimeters before the MOT. These beams are distinct from the MOT light and are aligned to cross the atomic beam without hitting the trapped cloud, allowing the Zeeman slower to be run at a higher final velocity and shortening the vulnerable free-flight time. The paper quantifies the problem with the estimate σ ≈ 2 d v_trans / v_long, where d is the free-flight distance, showing that for Dy the transverse spread is an order of magnitude larger than the MOT beam diameter.

What would settle it

Load a MOT with the atomic beam blocked, turn on the angled slowing beams at 7 mW each with the stated alignment, and measure the trapped-atom loss rate and any fluorescence: if the beams hit trapped atoms appreciably, the enhancement mechanism would be partly direct cooling or repumping rather than beam slowing alone. Separately, move the angled-beam intersection point upstream by several centimeters and check whether the population gain drops as the free-flight-time argument predicts.

Watch

Extended reading notes

Core claim

The central claim is that the free-flight time of the slowed beam, not the total slowed flux, is the limiting resource for loading a narrow-line Dy MOT. With an increasing-field Zeeman slower and a 16 cm free-flight distance, atoms exiting at the nominal capture velocity spread to a transverse size much larger than the MOT beams (estimated ~12 cm versus 2 cm), so most are lost. The paper's solution is to let atoms leave the Zeeman slower at a velocity above the MOT capture velocity and then add a final slowing stage: a pair of red-detuned 421 nm beams (optimal detuning −50 MHz, 7 mW per beam, ~5 mm diameter) that intersect the atomic beam directly in front of the MOT. Their transverse scattering forces cancel while their longitudinal components add, effectively raising the capture velocity and reducing the distance atoms must travel slowly. With this scheme the MOT loads about 3×$10^{8}$ atoms in 2 seconds, a factor of more than 20 above the best result without angled slowing.

Load-bearing premise

The claim that the population gain comes from reduced free-flight time assumes the angled slowing beams do not themselves scatter light from atoms already trapped in the MOT; the paper aligns the beams to miss the MOT but reports no direct measurement that the trapped atoms are unaffected.

Editorial extensions

If this is right

Without any change to the Zeeman slower, adding a few milliwatts per angled beam turns a 10^7-atom Dy MOT into a 3×10^8-atom MOT with 2 s loading, making high-number Dy samples routinely available for evaporation into an optical dipole trap.
Because the benefit comes from shortening free-flight time, experiments with long slow-to-MOT distances or with increasing-field slowers (which require compensation coils near the MOT) should see the largest gains.
The same pair-of-beams geometry should work for other narrow-line lanthanide species such as erbium and ytterbium, where capture velocities are similarly low.
The observation that the optimal bias field is independent of angled-beam detuning indicates the slowed velocity distribution is broad, so the enhancement does not rely on finely tuned velocity matching and should be reproducible without delicate frequency control.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A direct test of the proposed mechanism would be to vary the free-flight distance (or the position of the angled beams) and check that the population gain scales with the reduction in free-flight time; if most of the gain were instead direct cooling from stray beam light on the MOT, the gain would be insensitive to beam position.
The paper's scaling estimate σ ≈ 2 d v_trans/v_long is a ready-made design rule: any experiment can predict its expected gain by comparing this spread to its MOT beam diameter before installing angled slowing.
An interesting extension, not explored in the paper, would be to use the angled beams in a pulsed or chirped manner, which might capture a larger fraction of the broad velocity distribution at even lower average power than the continuous red detuning used here.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Dear colleague,

If you care about loading narrow-line MOTs of lanthanide atoms, read the setup section of this paper. The authors take the angled-slowing idea from Plotkin-Swing's Yb thesis and apply it to Dy, reporting a >20x gain in MOT population, reaching ~3e8 atoms in 2 s on the 626 nm transition. Even if the exact factor is softer than claimed, the core observation is real and practically useful.

What's solid: the central measurement is simple and direct—MOT population with and without the slowing arrangement, with a clear loading curve in Fig. 2. The optimization scans in Fig. 3 vary power and detuning of the 421 nm angled beams and show a sensible broad optimum. The estimates in Sec. III (capture velocity, transverse spread) are order-of-magnitude physics and labeled as such. Citation-wise, they credit the Yb origin properly; I see no self-citation inflation.

The soft spot is the baseline. Fig. 2 compares angled slowing + transverse cooling on against both off. The text's baseline for 'without angled slowing' is 'optimization of our Zeeman slowing parameters led to about 10^7 atoms,' but it never says whether transverse cooling was on for that number. If the 10^7 number had transverse cooling on, then the Fig. 2 orange curve is not the right baseline and the claimed factor of 20 is not directly evidenced. If it had transverse cooling off, the gain includes the known benefit of transverse cooling, so the incremental contribution of angled slowing is not isolated. This is a missing control, not a suspicious result—the angled beams clearly help—but the headline magnitude needs a fix. A simple figure with angled slowing on/off at fixed transverse cooling would settle it.

The residual-scattering point (do the angled beams hit the MOT?) is real but secondary, since the empirical enhancement would stand even if some scattering occurred.

Verdict: worth a serious referee, and likely publishable after the control is added. For anyone building a Dy/Er/Yb machine, this is a useful technical data point; for the rest, a quick skim is enough.

Referee Report

2 major / 5 minor

Summary. The paper reports a two-stage slowing scheme for loading a narrow-line Dy magneto-optical trap (MOT). In the first stage, atoms are Zeeman-slowed to a velocity above the MOT capture velocity; in the second stage, a pair of near-detuned 421 nm beams intersect the atomic beam directly in front of the MOT to slow atoms into the capture range. The authors argue that this 'angled slowing' reduces the transverse spreading of the slowed beam during free flight, and they report a more than factor-of-20 enhancement in final MOT population, reaching about 3e8 atoms in 2 s. The paper includes simple estimates of capture velocity and beam spread (Eqs. 1 and 2), an optimization of the angled-slowing power and detuning (Fig. 3), and measurements of the compressed MOT temperature and phase-space density.

Significance. If the reported enhancement is real, this is a useful and technically simple addition to narrow-line MOT experiments for Dy and similar species. The physical estimates are transparent, the beam-power requirement is low (7 mW per beam), and the geometry is much simpler than two-stage or core-shell MOT alternatives. The direct population measurements and optimization scans are the core evidence, and the paper's claimed mechanism is physically plausible. However, the quantitative headline claim ('factor of 20') depends on a baseline condition that is not unambiguously specified, and Figure 2 does not isolate the angled-slowing effect from transverse cooling. This must be clarified before the claim can be fully trusted.

major comments (2)

[Section III and Figure 2] The baseline for the factor-of-20 claim is not unambiguously defined. The text states 'Without employing angled slowing, optimization of our Zeeman slowing parameters led to a MOT population of about 10^7 atoms,' but does not state whether the transverse cooling beams were on. Figure 2's orange curve is explicitly 'both transverse cooling and angled slowing beams were turned off,' so it does not isolate the effect of angled slowing. If the 10^7 baseline included transverse cooling, the correct comparison for the angled-slowing gain is missing from Figure 2; if the 10^7 baseline excluded transverse cooling, the factor of 20 conflates angled slowing with the known benefit of transverse cooling. Please state the transverse-cooling status of the 10^7 baseline and provide a direct comparison with angled slowing off but transverse cooling on, or rephrase the quantitative claim to match the comparison actually shown.
[Section IV and Figure 3] The optimization scans of angled-slowing power and detuning (Figure 3) do not include a zero-power or beams-blocked point, so they do not by themselves establish the incremental gain over the standard Zeeman-slowing configuration. The text's 10^7-atom baseline is a useful reference, but it is not tied to the conditions of Figure 3. Please add a reference condition with the angled-slowing beams off (or otherwise explicitly cross-reference the baseline) so that the enhancement can be read directly from the optimization data.

minor comments (5)

[Figure 2 caption] The caption says the orange curve shows 'both transverse cooling and angled slowing beams turned off,' while the text says 'Figure 2 shows the population with and without angled slowing.' Please reconcile this wording so the reader knows exactly what the comparison is.
[Abstract and Section VI] The abstract and conclusion state 'more than an order of magnitude enhancement,' while the introduction and Section III state 'more than a factor of 20.' Please unify these numbers or explain the relationship between them.
[Equation (1)] The typeset form of vcap is ambiguous; as written it could be read as sqrt(2 hbar k Gamma / (2m) * D) or sqrt(2 hbar k Gamma / (2mD)). The numerical values in the text indicate the former is intended, but please write the formula explicitly as vcap = sqrt(hbar k Gamma D / m) or add parentheses.
[Section III] The interpretation that the enhancement comes from reduced free-flight time relies on the angled-slowing beams not scattering atoms already in the MOT. The paper states the beams are aligned to avoid the MOT, but no measurement of residual scattering is reported. A brief check (e.g., angled beams blocked versus unblocked with the atomic beam blocked) would make the mechanism claim more robust.
[Figures 2 and 3] The population data in Figures 2 and 3 are presented without error bars or a statement of shot-to-shot reproducibility. Adding representative error bars or at least noting the reproducibility would strengthen the quantitative claims.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the claimed enhancement is a direct measured comparison, and the supporting estimates use fixed constants without being fitted to the outcome.

full rationale

This paper is an experimental report. Its central claim that a two-stage 'angled slowing' scheme increases the final MOT population by more than a factor of 20 rests on a direct comparison of measured atom numbers (Fig. 2, Fig. 3, Sec. IV), not on a derived quantity that is fed back as an input. The capture-velocity estimate (Eq. 1) and beam-spread estimate (Eq. 2) are order-of-magnitude physical estimates computed from fixed constants (m, hbar, k, Gamma, D, and measured or typical velocities) and are used only to motivate why transverse spreading matters for a narrow-line Dy MOT; they are not fit to the observed enhancement and do not by construction produce the factor of 20. The scheme's prior introduction is attributed to an independent Yb experiment [14], and no uniqueness theorem or load-bearing self-citation is invoked. The baseline ambiguity noted in the skeptical summary (whether the 'no angled slowing' baseline had transverse cooling on) is a reproducibility or interpretation concern, not a circularity: the enhancement ratio is still an empirical comparison, and no equation in the paper reduces to its own input. Therefore the derivation chain is self-contained with respect to circularity.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claim is empirical, so the axiom ledger is short. The main assumptions are standard laser-cooling physics and geometric assumptions about the angled beams. No free parameters are fitted to the reported enhancement; the experimental control settings (detuning, power) are optimized but are not parameters of a model that claims to predict the enhancement.

assumptions (4)

domain assumption The transverse velocity distribution of the atomic beam has an RMS speed around 1% of the average longitudinal velocity (based on ref 18).
Used in Eq. (2) to estimate beam spread and motivate the need for angled slowing; based on prior work, not verified in this experiment.
domain assumption The two angled slowing beams have oppositely oriented transverse scattering forces that cancel exactly, providing only longitudinal slowing.
Assumes perfect symmetry in intensity and alignment; if transverse forces do not cancel, the beams would impart transverse heating or deflection.
domain assumption The angled slowing beams do not interact with atoms already trapped in the MOT.
The beams are aligned to intersect the atomic beam before the MOT, but no direct measurement of MOT light scattering from these beams is reported. This is load-bearing for attributing the population increase to enhanced loading rather than direct MOT cooling.
standard math Standard scattering force F = hbar*k*Gamma/2 for a saturated two-level transition (used in Eq. 1 for capture velocity).
Standard result in laser cooling; used to estimate capture velocity for Rb and Dy.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enhancing the capture velocity of a Dy magneto-optical trap with two-stage slowing." pith.science (2026). https://pith.science/paper/26EGL75F

@misc{pith2026190810433,
  author       = {Pith},
  title        = {Pith review of: Enhancing the capture velocity of a Dy magneto-optical trap with two-stage slowing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/26EGL75F}},
  note         = {Machine review of arXiv:1908.10433}
}

abstract

Magneto-optical traps (MOTs) based on the $626\;{\rm nm}$, $136\;{\rm kHz}$-wide intercombination line of Dy, which has an attractively low Doppler temperature of $3.3\;\mu{\rm K}$, have been implemented in a growing number of experiments over the last several years. A challenge in loading these MOTs comes from their low capture velocities. Slowed atomic beams can spread out significantly during free-flight from the Zeeman slower to the MOT position, reducing the fraction of the beam captured by the MOT. Here we apply, for the first time in a Dy experiment, a scheme for enhancing the loading rate of the MOT wherein atoms are Zeeman-slowed to a final velocity larger than the MOT's capture velocity, and then undergo a final stage of slowing by a pair of near-detuned beams addressing the $421\;{\rm nm}$ transition directly in front of the MOT. By reducing the free-flight time of the Zeeman-slowed atomic beam, we greatly enhance the slowed flux delivered to the MOT, leading to more than an order of magnitude enhancement in the final MOT population.

Figures

Figures reproduced from arXiv: 1908.10433 by the authors.

**Figure 2.** FIG. 2. Population of the MOT as a function of loading time. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** FIG. 3. Optimizing angled slowing. Dependence of the MOT [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 18 canonical work pages

[1]

M. Lu, N. Q. Burdick, S. H. Youn, and B. L. Lev, Strongly dipolar bose-einstein conden- sate of dysprosium, PRL 107, 190401 (2011), 6 DOI:10.1103/PhysRevLett.107.190401

doi:10.1103/physrevlett.107.190401 2011
[2]

Maier, M

T. Maier, M. S. H. Kadau, A. Griesmaier, and T. Pfau, Narrow-line magneto-optical trap for dys- prosium atoms, Optics Letters 39, 3138 (2014), DOI:10.1364/OL.39.003138

work page doi:10.1364/ol.39.003138 2014
[3]

Lucioni, G

E. Lucioni, G. Masella, A. Fregosi, C. Gabbanini, S. Gozzini, A. Fioretti, L. Del Bino, J. Catani, G. Mod- ugno, and M. Inguscio, A new setup for experiments with ultracold dysprosium atoms, The European Physi- cal Journal Special Topics 226, 2775 (2017)

work page 2017
[4]

Dreon, L

D. Dreon, L. A. Sidorenkov, C. Bouazza, W. Maineult, J. Dalibard, and S. Nascimbene, Optical cooling and trapping of highly magnetic atoms: the beneﬁts of a spontaneous spin polarization, Journal of Physics B: Atomic, Molecular and Optical Physics 50, 065005 (2017)

work page 2017
[5]

M¨ uhlbauer, N

F. M¨ uhlbauer, N. Petersen, C. Baumg¨ artner, L. Maske, and P. Windpassinger, Systematic optimization of laser cooling of dysprosium, Applied Physics B 124 (2018)

work page 2018
[6]

Ilzhfer, G

P. Ilzhfer, G. Durastante, A. Patscheider, A. Trautmann, M. J. Mark, , and F. Ferlaino, Two-species ﬁve-beam magneto-optical trap for erbium and dysprosium, PRA 97, 023633 (2018)

work page 2018
[7]

Chomaz, D

L. Chomaz, D. Petter, P. Ilzhfer, G. Natale, A. Traut- mann, C. Politi, G. Durastante, R. van Bijnen, A. Patscheider, M. Sohmen, M. Mark, and F. Ferlaino, Long-lived and transient supersolid behaviors in dipolar quantum gases, PRX 9, 021012 (2019)

work page 2019
[8]

Tanzi, E

L. Tanzi, E. Lucioni, F. Fam, J. Catani, A. Fioretti, C. Gabbanini, R. Bisset, L. Santos, , and G. Modugno, Observation of a dipolar quantum gas with metastable supersolid properties, PRL 122, 130405 (2019)

work page 2019

Show all 19 references

[9]

Ferrier-Barbut, H

I. Ferrier-Barbut, H. Kadau, M. Schmitt, M. Wenzel, and T. Pfau, Observation of quantum droplets in a strongly dipolar bose gas, PRL 116, 215301 (2016)

2016
[10]

Bttcher, J.-N

F. Bttcher, J.-N. Schmidt, M. Wenzel, J. Hertkorn, M. Guo, T. Langen, and T. Pfau, Transient supersolid properties in an array of dipolar quantum droplets, PRX 9, 011051 (2019)

2019
[11]

N. Q. Burdick, Y. Tang, and B. L. Lev, Long-lived spin-orbit-coupled degenerate dipolar fermi gas, PRX 6, 031022 (2016), DOI:10.1103/PhysRevX.6.031022

2016 doi
[12]

X. Cui, B. Lian, T.-L. Ho, B. L. Lev, and H. Zhai, Synthetic gauge ﬁeld with highly mag- netic lanthanide atoms, PRA 88, 011601 (2013), DOI:10.1103/PhysRevA.88.011601

2013 doi
[13]

S. H. Youn, M. Lu, U. Ray, and B. L. Lev, Dyspro- sium magneto-optical traps, PRA 82, 043425 (2010), DOI:10.1103/PhysRevA.82.043425

2010 doi
[14]

B. T. Plotkin-Swing, Large Momentum Separation Mat- ter Wave Interferometry , Ph.D. thesis, University of Washington (2018)

2018
[15]

J. Lee, J. H. Lee, J. Noh, and J. Mun, Core-shell magneto-optical trap for alkaline-earth-metal-like atoms, PRA 91, 053405 (2015)

2015
[16]

SVT Associates model SVTA-DF-20-450
[17]

Maier, Interactions in a Quantum Gas of Dysprosium Atoms, Ph.D

T. Maier, Interactions in a Quantum Gas of Dysprosium Atoms, Ph.D. thesis, Universitt Stuttgart (2015)

2015
[18]

When transverse Doppler cooling of the atomic beam is employed prior to Zeeman slowing, the transverse veloc- ity distribution can become much narrower; however, the light scattering during Zeeman slowing causes the trans- verse velocity distribution to re-broaden
[19]

Lu, Quantum Bose and Fermi Gases of Dysprosium: Production and Initial Study , Ph.D

M. Lu, Quantum Bose and Fermi Gases of Dysprosium: Production and Initial Study , Ph.D. thesis, Stanford Uni- versity (2014)

2014

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

M. Lu, N. Q. Burdick, S. H. Youn, and B. L. Lev, Strongly dipolar bose-einstein conden- sate of dysprosium, PRL 107, 190401 (2011), 6 DOI:10.1103/PhysRevLett.107.190401

doi:10.1103/physrevlett.107.190401 2011

[2] [2]

Maier, M

T. Maier, M. S. H. Kadau, A. Griesmaier, and T. Pfau, Narrow-line magneto-optical trap for dys- prosium atoms, Optics Letters 39, 3138 (2014), DOI:10.1364/OL.39.003138

work page doi:10.1364/ol.39.003138 2014

[3] [3]

Lucioni, G

E. Lucioni, G. Masella, A. Fregosi, C. Gabbanini, S. Gozzini, A. Fioretti, L. Del Bino, J. Catani, G. Mod- ugno, and M. Inguscio, A new setup for experiments with ultracold dysprosium atoms, The European Physi- cal Journal Special Topics 226, 2775 (2017)

work page 2017

[4] [4]

Dreon, L

D. Dreon, L. A. Sidorenkov, C. Bouazza, W. Maineult, J. Dalibard, and S. Nascimbene, Optical cooling and trapping of highly magnetic atoms: the beneﬁts of a spontaneous spin polarization, Journal of Physics B: Atomic, Molecular and Optical Physics 50, 065005 (2017)

work page 2017

[5] [5]

M¨ uhlbauer, N

F. M¨ uhlbauer, N. Petersen, C. Baumg¨ artner, L. Maske, and P. Windpassinger, Systematic optimization of laser cooling of dysprosium, Applied Physics B 124 (2018)

work page 2018

[6] [6]

Ilzhfer, G

P. Ilzhfer, G. Durastante, A. Patscheider, A. Trautmann, M. J. Mark, , and F. Ferlaino, Two-species ﬁve-beam magneto-optical trap for erbium and dysprosium, PRA 97, 023633 (2018)

work page 2018

[7] [7]

Chomaz, D

L. Chomaz, D. Petter, P. Ilzhfer, G. Natale, A. Traut- mann, C. Politi, G. Durastante, R. van Bijnen, A. Patscheider, M. Sohmen, M. Mark, and F. Ferlaino, Long-lived and transient supersolid behaviors in dipolar quantum gases, PRX 9, 021012 (2019)

work page 2019

[8] [8]

Tanzi, E

L. Tanzi, E. Lucioni, F. Fam, J. Catani, A. Fioretti, C. Gabbanini, R. Bisset, L. Santos, , and G. Modugno, Observation of a dipolar quantum gas with metastable supersolid properties, PRL 122, 130405 (2019)

work page 2019