Loss mechanisms, heating, filling, and lifetime limits for long-lived ground-state polar molecules in a 3D optical lattice

Ultracold fermionic KRb molecules in a 3D optical lattice can have single-particle lifetimes of up to 25 s once chemical reactions are frozen out by site localization, but Chotia et al. (Phys. Rev. Lett. 108, 080405 (2012)) left a specific set of issues open: the unknown origin of a fast initial loss of molecules in the first hundreds of milliseconds after STIRAP and lattice loading; the unidentified microscopic heating sources that limit lifetimes in the weak-lattice regime; the difficulty of achieving high global lattice filling (many sites with exactly one K and one Rb atom) for a heteronuclear Bose-Fermi mixture; the absence of a systematic theory-experiment account of many-body dipolar dynamics, rotational spectroscopy, and coherence in partially filled lattices; and the off-resonant photon-scattering ceiling on trap lifetimes set by the imaginary part of the molecular polarizability.

Background

Ultracold bialkali molecules such as fermionic $^{40}$K$^{87}$Rb (KRb) are produced by associating ultracold K and Rb atoms across a magnetic-field Feshbach resonance (a scattering resonance that binds the two atoms into a weakly bound "Feshbach molecule") and then coherently transferring these to the rovibrational ground state using STIRAP (stimulated Raman adiabatic passage, a two-photon optical pulse sequence). Ground-state KRb carries a permanent electric dipole moment, so a gas of such molecules interacts through long-range, anisotropic dipole-dipole interactions, which is the basis of proposals to simulate lattice spin models and strongly correlated many-body physics. The historical obstacle was lifetime: KRb molecules undergo exothermic bimolecular chemical reactions (KRb + KRb giving $K_2$ + $Rb_2$) even at ultracold temperatures, and applying an electric field to polarize the molecules increases the reaction rate further through the attractive part of the dipole-dipole interaction. Confining the molecules in a 3D optical lattice, a periodic array of microscopic traps formed by interfering laser beams with depths quoted in units of the molecular recoil energy $E_R$ (the kinetic energy gained by absorbing one lattice photon), localizes individual molecules on separate sites and suppresses these collisions. In this setting Chotia et al. (Phys. Rev. Lett. 108, 080405 (2012)) demonstrated $1/e$ lifetimes of 16.3(1.5) s and up to 25 s for ground-state KRb in a 3D lattice, with no measurable lifetime reduction for a modest induced dipole moment of 0.17 Debye (1 Debye $\approx 3.34\times10^{-30}$ C$\cdot$m), and lifetimes longer than 20 s for weakly bound Feshbach molecules, a 100-fold improvement. Once reactions are frozen out, the lifetime ceiling is set by off-resonant scattering of the trapping light: a molecule's dense electronic spectrum makes most scattered photons eject it from its internal state, so the loss rate is proportional to the imaginary part of the molecular ac polarizability at the trapping wavelength, which the authors measured at 1064 nm. Within this single experimental setting, the authors left five issues unresolved, which jointly gate one objective, a long-lived, low-entropy, coherently interacting dipolar gas in the lattice whose residual loss mechanisms are microscopically understood. First, a rapid initial loss of molecules in the first few hundreds of milliseconds after STIRAP and lattice loading remained unexplained, tentatively attributed to collisions with impurities such as molecules in excited internal states produced during the STIRAP transfer. Second, in the weak-lattice regime (a small lattice potential added along tube-shaped traps), the data were consistent with heating, plausibly optical phase noise of the lattice beams appearing as translational lattice noise that promotes molecules to higher bands where they are mobile and can collide, but this was modeled only phenomenologically with a constant heating rate of 1-2 $E_R$/s. Third, only lattice sites containing exactly one K and one Rb atom can be converted into molecules, and creating many such pre-formed pairs in a heteronuclear Bose-Fermi mixture was an unsolved loading problem, limiting global lattice filling, molecule production, and entropy. Fourth, while isolated molecules showed long single-body lifetimes, a systematic theoretical and experimental understanding of many-body dipolar dynamics, rotational spectroscopy responses, and coherence in partially filled lattices, where dipolar interactions between neighboring sites have energies on the order of a few hundred Hz, remained to be developed. Fifth, off-resonant photon scattering set the upper bound on lifetimes in deep traps, so extending coherence and hold times further would require trapping wavelengths, intensities, or state-insensitive trapping schemes that reduce the imaginary part of the molecular polarizability.

Later work has partially addressed several of these issues. The loading problem of the third issue has been substantially advanced by loading a bosonic Rb Mott insulator together with a fermionic K band insulator and converting the pre-formed atom pairs into ground-state molecules, reaching about 25% filling with an entropy of 2.2 $k_B$ per molecule (Moses et al. 2015), while degenerate molecular Fermi gases have since been produced by dipolar evaporation (Lin et al. 2026). The fourth issue has become an active experimental program: dipolar spin-exchange interactions were observed for lattice-confined KRb (Yan et al. 2013), and tunable dipolar many-body dynamics and second-scale rotational coherence have been realized in later experiments (Miller et al. 2024; Carroll et al. 2025; Gregory et al. 2024). The photon-scattering ceiling of the fifth issue has been mitigated for other bialkali species through magic and tune-out trapping (Seeßelberg et al. 2018; Bause et al. 2020), but off-resonant light scattering still dominates the one-body loss of KRb in deep 3D lattices. By contrast, the microscopic origin of the fast initial loss and the microscopic sources of the weak-lattice heating remain unidentified for lattice-trapped molecules: no later work establishes the dominant channels, and recent experiments still describe number loss with phenomenological one-body and two-body rate models.

Problem Statement

For ground-state and Feshbach KRb molecules confined in a 3D optical lattice, resolve the issues left open by the demonstration of long (up to 25 s) single-body lifetimes, all in service of one objective, a long-lived, low-entropy, coherently interacting gas of dipolar molecules in the lattice whose residual loss mechanisms are microscopically understood rather than phenomenologically modeled: (1) determine the origin of the rapid initial loss of molecules on timescales of hundreds of milliseconds following STIRAP and lattice loading, including whether it arises from impurities such as molecules in excited internal states produced during STIRAP, from residual unpaired atoms, or from another mechanism, together with the corresponding rates and mitigation strategies; (2) identify the microscopic sources of the heating that limits molecule lifetimes in the weak-lattice regime (for example, optical phase noise of the lattice beams producing translational lattice noise that promotes molecules to higher bands), determine their dependence on lattice beam parameters, and establish methods to suppress this heating so as to extend lifetimes at intermediate lattice depths; (3) develop loading and optimization procedures, including exploitation of interspecies interaction effects, that raise the global lattice filling, the fraction of sites occupied by exactly one K and one Rb atom, and thereby lower the entropy and increase molecule production of the heteronuclear Bose-Fermi mixture; (4) build a systematic theoretical and experimental account of many-body dipolar dynamics, rotational spectroscopy responses, and coherence in partially filled lattices, including how dipolar interactions between neighboring sites, with interaction energies on the order of a few hundred Hz, modify observables and decoherence; and (5) identify trapping wavelengths, intensities, or state-insensitive trapping schemes that reduce the imaginary part of the molecular polarizability, so that coherence and hold times in deep traps extend beyond the off-resonant photon-scattering limit.

The five parts all bear on one objective — a long-lived, low-entropy, coherently interacting dipolar gas in the lattice whose residual loss mechanisms are microscopically understood — and each stands on its own as an unresolved scientific question. Parts (1) and (2) have no reported resolution; parts (3)-(5) are required only to the extent that they remain open, that is, their answers must extend beyond the partial results already obtained for KRb or in other bialkali species — higher lattice filling and lower entropy, a microscopic rather than phenomenological account of the many-body dipolar observables and decoherence, and KRb-specific trapping schemes that extend coherence and hold times beyond the present off-resonant photon-scattering limit.

References

1. A. Chotia, B. Neyenhuis, S. A. Moses, B. Yan, J. P. Covey, M. Foss-Feig, A. M. Rey, D. S. Jin, J. Ye, 'Long-Lived Dipolar Molecules and Feshbach Molecules in a 3D Optical Lattice', Phys. Rev. Lett. 108, 080405 (2012), DOI: 10.1103/PhysRevLett.108.080405, https://doi.org/10.1103/PhysRevLett.108.080405
2. S. A. Moses, J. P. Covey, M. T. Miecnikowski, B. Yan, B. Gadway, J. Ye, D. S. Jin, 'Creation of a low-entropy quantum gas of polar molecules in an optical lattice', Science 350, 659 (2015), DOI: 10.1126/science.aac6400, https://doi.org/10.1126/science.aac6400
3. B. Yan, S. A. Moses, B. Gadway, J. P. Covey, K. R. A. Hazzard, A. M. Rey, D. S. Jin, J. Ye, 'Observation of dipolar spin-exchange interactions with lattice-confined polar molecules', Nature 501, 521-525 (2013), DOI: 10.1038/nature12483, https://doi.org/10.1038/nature12483
4. J. Lin, J. He, M. Jin, G. Chen, D. Wang, 'Seconds-Scale Coherence on Nuclear Spin Transitions of Ultracold Polar Molecules in 3D Optical Lattices', Phys. Rev. Lett. 128, 223201 (2022), DOI: 10.1103/PhysRevLett.128.223201, https://doi.org/10.1103/PhysRevLett.128.223201
5. F. Seeßelberg, X.-Y. Luo, M. Li, R. Bause, S. Kotochigova, I. Bloch, C. Gohle, 'Extending Rotational Coherence of Interacting Polar Molecules in a Spin-Decoupled Magic Trap', Phys. Rev. Lett. 121, 253401 (2018), DOI: 10.1103/PhysRevLett.121.253401, https://doi.org/10.1103/PhysRevLett.121.253401
6. R. Bause, M. Li, A. Schindewolf, X.-Y. Chen, M. Duda, S. Kotochigova, I. Bloch, X.-Y. Luo, 'Tune-Out and Magic Wavelengths for Ground-State $^{23}$Na$^{40}$K Molecules', Phys. Rev. Lett. 125, 023201 (2020), DOI: 10.1103/PhysRevLett.125.023201, https://doi.org/10.1103/PhysRevLett.125.023201 (preprint arXiv:1912.10452, https://arxiv.org/abs/1912.10452)
7. Q. Guan, S. L. Cornish, S. Kotochigova, 'Magic conditions for multiple rotational states of bialkali molecules in optical lattices', Phys. Rev. A 103, 043311 (2021), DOI: 10.1103/PhysRevA.103.043311, https://doi.org/10.1103/PhysRevA.103.043311
8. P. D. Gregory, L. M. Fernley, A. L. Tao, S. L. Bromley, J. Stepp, Z. Zhang, S. Kotochigova, K. R. A. Hazzard, S. L. Cornish, 'Second-scale rotational coherence and dipolar interactions in a gas of ultracold polar molecules', Nature Physics 20, 415-421 (2024), DOI: 10.1038/s41567-023-02328-5, https://doi.org/10.1038/s41567-023-02328-5 (preprint arXiv:2306.02991, https://arxiv.org/abs/2306.02991)
9. A. N. Carroll, H. Hirzler, C. Miller, D. Wellnitz, S. R. Muleady, J. Lin, K. P. Zamarski, R. R. W. Wang, J. L. Bohn, A. M. Rey, J. Ye, 'Observation of generalized t-J spin dynamics with tunable dipolar interactions', Science 388, 381-386 (2025), DOI: 10.1126/science.adq0911, https://doi.org/10.1126/science.adq0911
10. C. Miller, A. N. Carroll, J. Lin, H. Hirzler, H. Gao, H. Zhou, M. D. Lukin, J. Ye, 'Two-axis twisting using Floquet-engineered XYZ spin models with polar molecules', Nature 633, 332-337 (2024), DOI: 10.1038/s41586-024-07883-2, https://doi.org/10.1038/s41586-024-07883-2
11. J. P. Nohra, C. Daix, J. Verstraten, M. Dixmerias, T. de Jongh, B. Peaudecerf, F. Gerbier, T. Yefsah, 'Phase-Noise-Induced Heating in Optical Lattices', arXiv:2608.07442 (2026), DOI: 10.48550/arXiv.2608.07442, https://arxiv.org/abs/2608.07442
12. J. Lin, A. N. Carroll, P. Martin, C. Miller, R. R. W. Wang, K. Xu, J. L. Bohn, T. de Jongh, J. Ye, 'Fermi gas of polar molecules in the Pauli-blocked regime', arXiv:2606.14135 (2026), DOI: 10.48550/arXiv.2606.14135, https://arxiv.org/abs/2606.14135