A 2024 experiment (Chen et al., Laser & Photonics Reviews 19(6), 2401595) demonstrated a multifunctional metalens that focuses an 852 nm trapping beam into a circularly polarized single-atom tweezer while collecting 780 nm fluorescence from the same 87^{87}Rb atom. In that tweezer the trap lifetime changes by an order of magnitude when the axial bias magnetic field BzB_z is tuned by only about 0.1 G, and the optimal BzB_z extrapolates to a nonzero value at zero trapping power, violating the Bz+BF=0B_z+B_F=0 cancellation expected from the established polarization-gradient-cooling model, where BFB_F is the light-induced fictitious magnetic field of the circularly polarized trap. The authors explicitly state that this drastic change "cannot be explained by the resonance effect and warrants further investigations in the future". The open problem is to identify and experimentally validate the microscopic mechanism (or mundane artifact) behind the BzB_z-dependent lifetime and the breakdown of the Bz+BF=0B_z+B_F=0 expectation, accounting for the actual non-Gaussian focal field of the metalens.

Background

An optical tweezer is a tightly focused far-off-resonant laser beam whose dipole force traps a single neutral atom at its waist; neutral-atom tweezer arrays are a leading platform for quantum simulation, quantum information processing, and precision sensing. Conventional tweezer setups require bulky high-numerical-aperture (NA) microscope objectives plus separate wave plates and dichroic optics for trapping and fluorescence collection. A metalens is a flat, subwavelength-patterned dielectric surface (a metasurface) that shapes the phase, amplitude, and polarization of light to replace such bulk optics; Chen et al. (Laser & Photonics Reviews 19(6), 2401595; arXiv:2411.05501) fabricated one that simultaneously acts as an achromatic lens (focusing 852 nm trapping light and 780 nm fluorescence light from a 87^{87}Rb atom to the same point), a quarter-wave plate (converting the trap beam to circular polarization), and a circular polarizer, all on a sapphire chip inside a vacuum cell. When circularly polarized light traps an alkali atom, the vector part of the ac Stark shift (the difference of the light shifts of the Zeeman sublevels) acts on the atomic spin like an effective magnetic field along the beam axis - the fictitious magnetic field BFB_F, proportional to the local trap intensity. Earlier single-atom work with a metasurface-lens tweezer (Hsu et al., PRX Quantum 3, 030316 (2022)) used linear polarization; the metalens of Chen et al. is the first to combine circular polarization with dual-wavelength trapping and readout. In a circularly polarized tweezer, polarization-gradient cooling (PGC, the standard sub-Doppler cooling mechanism in which spatially varying light polarization creates Sisyphus-type friction) of a single atom is known to be very sensitive to trap-beam ellipticity: Chin, Steiner and Kurtsiefer (Physical Review A 96, 033406 (2017)) found that σ+−σ−\sigma^+-\sigma^- PGC of a single atom is inhibited by BFB_F, so that the longest lifetime under probe illumination should occur when an external axial bias field cancels the fictitious field, Bz+BF=0B_z+B_F=0. Chen et al. observed instead a sharp resonance-like lifetime feature: at trap powers of 14.0-18.6 mW the lifetime rises from about 100 ms at Bz=0.48B_z=0.48 G and 0.68 G to about 1 s at 0.59 G, i.e. an order-of-magnitude change over ∼0.1\sim0.1 G. They ruled out dark states (the atomic fluorescence emission rate itself is insensitive to BzB_z; the reduced count rates at short lifetimes track atom escape rather than reduced emission) and simple resonance effects, confirmed that the optimal BzB_z grows with trap power (consistent with a growing BFB_F), but found that a linear fit of the optimal BzB_z versus power extrapolates to Bz≠0B_z\neq0 at P=0P=0, violating the Bz+BF=0B_z+B_F=0 prediction. Candidate missing physics they list includes additional light shifts, vectorial focusing effects, substrate-induced fields, and spin-motion coupling from the strong fictitious-field gradient ∼BF/w0\sim B_F/w_0 across the w0=1.33 μw_0=1.33\,\mum waist, which no uniform bias field can compensate. Any explanation must also contend with the measured focal field: the metalens focus is non-Gaussian (measured Rayleigh length 11.68 μ\mum versus 6.52 μ\mum expected for a Gaussian of the same waist, with side lobes and only ζ≈33%\zeta\approx33\% of the incident power in the central peak), the substrate sapphire is birefringent, and the device was designed for flat-top rather than the experimental Gaussian illumination. The broader field has advanced rapidly on scalability - metasurface-generated tweezer arrays with over 1000 trapped strontium atoms (Holman et al., Nature 649, 859 (2026)), 78,400-site tweezer arrays (Wang et al., Chin. Phys. Lett. 43, 010606), and 11,000 trapped atoms from a single 2-cm metasurface (Wang et al., 2026) - and a May 2026 review (Fang et al., arXiv:2605.30498) confirms that efficiency and tunability of such multifunctional metasurfaces remain open engineering challenges, but no work has addressed the anomalous magnetic-field-dependent trap dynamics.

Problem Statement

Explain, with a validated microscopic model or a decisive experiment, the bias-magnetic-field dependence of the single-atom lifetime in the circularly polarized optical tweezer generated by the multifunctional metalens of Chen et al. (Laser & Photonics Reviews 19(6), 2401595). Specifically, an admissible answer must account for, within one coherent mechanism or artifact identification (a coherent account may combine more than one coexisting effect): (i) the order-of-magnitude change of the 87^{87}Rb trap lifetime over an axial bias-field change of only ∼0.1\sim0.1 G (from about 100 ms at Bz=0.48B_z=0.48 G and 0.68 G to about 1 s at 0.59 G, at trapping powers 14.0, 16.3, and 18.6 mW), while the atomic fluorescence emission rate itself stays insensitive to BzB_z (the reduced count rates at short lifetimes track atom escape rather than reduced emission); (ii) why the optimal BzB_z increases with trapping power yet extrapolates to a nonzero intercept at P=0P=0, in violation of the Bz+BF=0B_z+B_F=0 cancellation expected from the polarization-gradient-cooling model of Chin, Steiner and Kurtsiefer (Physical Review A 96, 033406 (2017)), where BFB_F is the intensity-proportional fictitious magnetic field of the circularly polarized trap; and (iii) which candidate effect dominates - vector light shifts of the tightly focused, vectorial focal field; spin-motion coupling induced by the fictitious-field gradient ∼BF/w0\sim B_F/w_0 that no uniform bias field can compensate; residual polarization ellipticity from the birefringent sapphire substrate; substrate-induced or stray fields; or an experimental artifact. An admissible answer must be evaluated against the actually measured focal field of the metalens (non-Gaussian axial energy distribution with Rayleigh length 11.68 µm, side lobes, and ζ≈33%\zeta\approx33\% axial concentration) rather than an ideal Gaussian mode, and must reproduce the published lifetime-versus-BzB_z behavior at the measured powers with a validated model or specify a decisive experiment that discriminates among the candidate mechanisms.

References

  1. Guang-Jie Chen, Dong Zhao, Zhu-Bo Wang, Ziqin Li, Ji-Zhe Zhang, Liang Chen, Yan-Lei Zhang, Xin-Biao Xu, Ai-Ping Liu, Chun-Hua Dong, Guang-Can Guo, Kun Huang, Chang-Ling Zou, 'Multifunctional Metalens for Trapping and Characterizing Single Atoms', Laser & Photonics Reviews 19(6), 2401595, 2024 (online 24 December 2024; print issue March 2025). DOI: 10.1002/lpor.202401595, https://doi.org/10.1002/lpor.202401595; arXiv:2411.05501, https://arxiv.org/abs/2411.05501
  2. Yue-Sum Chin, Matthias Steiner, Christian Kurtsiefer, 'Polarization gradient cooling of single atoms in optical dipole traps', Physical Review A 96, 033406, 2017. DOI: 10.1103/PhysRevA.96.033406, https://doi.org/10.1103/PhysRevA.96.033406; arXiv:1707.05965, https://arxiv.org/abs/1707.05965
  3. Ting-Wei Hsu, Wenqi Zhu, Tobias Thiele, Mark O. Brown, Scott B. Papp, Amit Agrawal, Cindy A. Regal, 'Single-Atom Trapping in a Metasurface-Lens Optical Tweezer', PRX Quantum 3, 030316, 2022. DOI: 10.1103/PRXQuantum.3.030316, https://doi.org/10.1103/PRXQuantum.3.030316; arXiv:2110.11559, https://arxiv.org/abs/2110.11559
  4. Aaron Holman, Yuan Xu, Ximo Sun, Jiahao Wu, Mingxuan Wang, Zezheng Zhu, Bojeong Seo, Nanfang Yu, Sebastian Will, 'Trapping of single atoms in metasurface optical tweezer arrays', Nature 649, 859-865, 2026. DOI: 10.1038/s41586-025-09961-5, https://doi.org/10.1038/s41586-025-09961-5; arXiv:2411.05321, https://arxiv.org/abs/2411.05321
  5. Chengyu Fang, Minjeong Kim, Mark Saffman, Jennifer T. Choy, Mikhail Kats, 'Metasurfaces for neutral-atom trapping', 2026. arXiv:2605.30498, https://arxiv.org/abs/2605.30498
  6. Yuqing Wang, Zhongchi Zhang, Tao Zhang, Yuxuan Liao, Hanteng Wang, Ye Tian, Binjie Ji, Yujia Wu, Luming Ma, Chen Qing, Chengshu Li, Wei Zhang, Yidong Huang, Wenjun Zhang, Xue Feng, Wenlan Chen, Hui Zhai, 'Trapping 11,000 Atoms in a Tweezer Array Generated by a Single Metasurface', 2026. arXiv:2606.02715, https://arxiv.org/abs/2606.02715
  7. Yuqing Wang, Yuxuan Liao, Tao Zhang, Ye Tian, Yujia Wu, Wenjun Zhang, Wei Zhang, Yidong Huang, Hui Zhai, Wenlan Chen, Xue Feng, Zhongchi Zhang, 'Direct Generation of an Array with 78400 Optical Tweezers Using a Single Metasurface', Chinese Physics Letters 43, 010606, 2026. DOI: 10.1088/0256-307X/43/1/010606, https://doi.org/10.1088/0256-307X/43/1/010606; arXiv:2512.08222, https://arxiv.org/abs/2512.08222
  8. Chengyu Fang, Minjeong Kim, Hongyan Mei, Xuting Yang, Zhaoning Yu, Yuzhe Xiao, Sanket Deshpande, Preston Huft, Alan M. Dibos, David A. Czaplewski, Mark Saffman, Jennifer T. Choy, Mikhail A. Kats, 'Silicon-on-sapphire metasurfaces generate arrays of dark and bright traps for neutral atoms', 2026. arXiv:2601.01038, https://arxiv.org/abs/2601.01038
  9. Hae Lim, Johannes E. Fröch, Christian M. Pluchar, Arka Majumdar, Sara L. Mouradian, 'Scalable Ion Fluorescence Collection Using a Trap-Integrated Metalens', ACS Photonics 13, 2052-2058, 2026. DOI: 10.1021/acsphotonics.5c02393, https://doi.org/10.1021/acsphotonics.5c02393; arXiv:2507.23071, https://arxiv.org/abs/2507.23071
  10. Lei Xu, Ling-Xiao Wang, Guang-Jie Chen, Zhu-Bo Wang, Xin-Biao Xu, Guang-Can Guo, Chang-Ling Zou, Guo-Yong Xiang, 'Efficient single-atom transfer from an optical conveyor belt to a tightly confined optical tweezer', 2025. arXiv:2507.21456, https://arxiv.org/abs/2507.21456