Λ \Lambda -Enhanced Gray Molasses Cooling of 85 Rb Atoms in Tweezers Using the D 2 Line
D.A. Janse van Rensburg, R.C. Venderbosch, Y. van der Werf, J.J. del Pozo Mellado, M.L. Venderbosch, R.S. Lous, E.J.D. Vredenbregt, S.J.J.M.F. Kokkelmans · 2025
Read Λ \Lambda -Enhanced Gray Molasses Cooling of 85 Rb Atoms in Tweezers Using the D 2 Line by D.A. Janse van Rensburg, R.C. Venderbosch, Y. van der Werf, J.J. del Pozo Mellado, M.L. Venderbosch, R.S. Lous, E.J.D. Vredenbregt, S.J.J.M.F. Kokkelmans (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-17653.txt.
Introduction
Deon Janse van Rensburg and seven co-authors chill rubidium atoms held in laser tweezers down to 4.0 microkelvin, colder than the standard method reaches. Their trick is lambda-enhanced gray molasses, a laser cooling recipe that slows atoms by making them climb light hills, here applied to a rubidium line most labs avoid for this purpose. Colder atoms also hold their quantum state longer: the clock qubit coherence time grew by a factor of 1.5.
The setup needs no careful beam alignment and rides on the same lasers used to trap the atoms, needing only frequency and phase tweaks. A four-level computer model of the forces matched what the lab saw. Colder, longer-lived atoms are direct fuel for quantum computing and precision clocks.
Catalog record
| CREATOR | D.A. Janse van Rensburg, R.C. Venderbosch, Y. van der Werf, J.J. del Pozo Mellado, M.L. Venderbosch, R.S. Lous, E.J.D. Vredenbregt, S.J.J.M.F. Kokkelmans |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-17653
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Λ \Lambda -Enhanced Gray Molasses Cooling of 85 Rb Atoms in Tweezers Using the D 2 Line D.A. Janse van Rensburg, R.C. Venderbosch, Y. van der Werf, J.J. del Pozo Mellado, M.L. Venderbosch, R.S. Lous, E.J.D. Vredenbregt, S.J.J.M.F. Kokkelmans Abstract We demonstrate the implementation of Λ \Lambda -enhanced gray molasses cooling on the D 2 line of 85 Rb atoms in an optical tweezer array. This technique yields lower atomic temperatures of 4.0 ( 2 ) µ K 4.0(2)\text{\,}\mathrm{\SIUnitSymbolMicro K} compared to red-detuned polarization gradient cooling, and consequently extends the T 2 ∗ T_{2}^{*} coherence time of the hyperfine clock qubit by a factor of 1.5. The method is alignment-free and can be readily implemented on laser beams used for magneto-optical trapping, as it only requires frequency and phase modulation control. Our experimental observations are corroborated by a numerical model based on a semi-classical force approach extended to a four-level system, including two hyperfine states of the upper manifold that are 120 MHz 120\text{\,}\mathrm{MHz} apart. I Introduction Neutral atoms trapped in optical tweezers have shown promise as a platform for quantum simulation and quantum information experiments 23 ; 31 , with atoms of the alkali metal group commonly used for this purpose 2 ; 28 ; 21 . For those platforms, efficient sub-Doppler cooling is a requirement for achieving long coherence times, since the far red-detuned optical tweezers impose a differential light shift on the atoms. This results in dephasing that is temperature dependent for non-magic wavelength traps 20 ; 12 ; 38 . A versatile approach to reduce the temperature of single alkali atoms in optical tweezers – and thereby extend their coherence times – is \Lambda -enhanced gray molasses cooling ( \Lambda -GMC) 10 ; 9 . This scheme combines blue-detuned gray molasses cooling on a type-II transition ( F^{\prime}\leq F ) with velocity-selective coherent population trapping to minimize photon scattering events for the cold atomic population 11 . \Lambda -GMC is typically executed on the D 1 line due to the favorable three-level structure 32 ; 29 ; 4 ; 7 . \Lambda -GMC can also be implemented efficiently on the D 2 line, if the hyperfine splitting is much larger than the transition linewidth \Gamma . The technique has been successfully implemented on 87 Rb 30 , 133 Cs 14 and 40 K 3 alkali metal isotopes on the D 2 line, reaching temperatures of several \mathrm{\SIUnitSymbolMicro K} . In contrast, the small \sim 6\,\Gamma hyperfine splitting of 23 Na limits the achievable temperature reached by \Lambda -GMC on the D 2 line to 56\text{\,}\mathrm{\SIUnitSymbolMicro K} 33 , whereas 7\text{\,}\mathrm{\SIUnitSymbolMicro K} 7 has been achieved using the D 1 line. Still, the use of the D 2 line is convenient for most alkali atoms as the required laser is typically already used for a MOT, although it introduces the complication that blue-detuned light for the type-II F\rightarrow F^{\prime}=F transition is red-detuned from the type-I F\rightarrow F^{\prime}=F+1 transition. In this article, we demonstrate \Lambda -GMC on the D 2 line of 85 Rb, and explore the cooling efficiency in a four-level structure. This transition has a natural linewidth of \Gamma/2\pi=$6.066\,69(18)\text{\,}\mathrm{MHz}$ 34 , and features an intermediate hyperfine splitting of 19.9\,\Gamma between the type-I and type-II transitions. Furthermore, we demonstrate improved coherence times on the hyperfine clock qubit due to optimized \Lambda -GMC. To complement the experimental results, we have generalized the theoretical description of the \Lambda -GMC to a four-level system relevant to the D 2 line of alkali metal atoms. II \Lambda -GMC Demonstration II.1 Method We implement \Lambda -GMC in our neutral atom setup 36 ; 15 , which traps single atoms in a 10 \times 10 array of far red-detuned, 813\text{\,}\mathrm{nm} , optical tweezers. The core laser in the setup is 780\text{\,}\mathrm{nm} , which addresses the D 2 line of 85 Rb, as shown in the energy-level diagram in Fig. 1 . This laser is used for creating a magneto-optical trap, imaging single atoms, optical pumping into the 5 2 S {}_{1/2}|F=2,m_{F}=0\rangle state for qubit operations, qubit state projection via a hyperfine-selective blowout pulse, and \Lambda -GMC. We use an electro-optic modulator (EOM) to produce frequency sidebands for the 5^{2} S {}_{1/2}|F=2\rangle \leftrightarrow 5^{2} P {}_{3/2}|F^{\prime}=3\rangle repumper transition during the MOT stage, imaging of the single atoms, and \Lambda -GMC. The beam delivery system for \Lambda -GMC does not require hardware upgrades on the setup and is directly implemented on existing 3D MOT beams. During \Lambda -GMC, a repumper detuning \delta_{1} is tuned to the Raman resonance condition and can be scanned over a wide range. We operate an acousto-optic modulator (AOM) to scan the carrier detuning \delta_{2} over a range from +6\,\Gamma to +14.3\,\Gamma from the 5^{2} S {}_{1/2}|F=3\rangle \leftrightarrow 5^{2} P {}_{3/2}|F^{\prime}=3\rangle transition maintaining consistent laser power levels. The MOT is created at -3\,\Gamma from the free-space 5^{2} S {}_{1/2}|F=3\rangle \leftrightarrow 5^{2} P {}_{3/2}|F^{\prime}=4\rangle transition. We investigate the cooling behavior using the recapture probability after a 40\text{\,}\mathrm{\SIUnitSymbolMicro s} release time to study the relative heating and cooling of the atoms when addressing multiple transitions of the D 2 line. The initial temperature of the atoms before a 20\text{\,}\mathrm{ms} \Lambda -GMC pulse is 9.7(4)\text{\,}\mathrm{\SIUnitSymbolMicro K} determined by the release and recapture method 35 . F=2:|1\rangle F=3:|2\rangle 3.035 GHz F^{\prime}=1 F^{\prime}=2 F^{\prime}=3:|3\rangle F^{\prime}=4:|4\rangle \delta_{4}=120 MHz 5^{2}S_{1/2} 5^{2}P_{3/2} D_{2} 780.241 nm \Omega_{1} \Omega_{2} \delta_{1} \delta_{2} \delta Figure 1: D 2 line of 85 Rb, focusing on the four levels { |1\rangle,|2\rangle,|3\rangle,|4\rangle } involved in \Lambda -GMC. The parameters \{\delta_{1},\Omega_{1}\} represent the repumper detuning and Rabi frequency (blue, F=2\rightarrow F^{\prime}=3) , the repumper light is an EOM sideband on the cooling laser. The parameters \{\delta_{2},\Omega_{2}\} are the cooling laser carrier detuning and Rabi frequency (green, F=3\rightarrow F^{\prime}=3 ), \delta=\delta_{1}-\delta_{2} is the Raman detuning, and \delta_{4} the hyperfine splitting. II.2 Experimental Results As a first scan, we vary the repump detuning for fixed carrier detuning over the hyperfine manifold as shown in Fig. 2 a. We use a carrier detuning of \delta_{2}=6\,\Gamma with a Rabi frequency of \Omega_{2}=1.63\,\Gamma , and a repumper-to-carrier ratio of (\Omega_{1}/\Omega_{2})^{2}=0.1 . The typical recapture probability of the atoms at the initial temperature before cooling is indicated by the dashed black line. We observe heating when the light is red-detuned, and slight cooling when it is blue-detuned with respect to the S {}_{1/2}|F=2\rangle\leftrightarrow P {}_{3/2}|F^{\prime}=1\rangle and S {}_{1/2}|F=2\rangle\leftrightarrow P {}_{3/2}|F^{\prime}=2\rangle transitions. We attribute the heating to the fact that most scattering events occur on these ‘repumper’ transitions, which are type-II transitions where sub-Doppler cooling arises when the light is blue-detuned. This is similar to the condition that both the cooling and repumper light must be blue-detuned of their respective transitions to achieve the coldest temperatures in a blue-detuned MOT of alkali metal atoms 16 . Near resonance with the S {}_{1/2}|F=2\rangle\leftrightarrow P {}_{3/2}|F^{\prime}=3\rangle transition (\delta\approx-\delta_{2}) we observe pronounced cooling, which we attribute to gray molasses cooling. Finally, near the Raman resonance condition ( \delta=0 ) we observe the characteristic narrow linewidth cooling and heating feature of \Lambda -GMC. We perform similar measurements for various carrier and Raman detunings, shown in Fig. 2 b, to study the influence of the type-I S {}_{1/2}|F=3\rangle\leftrightarrow P {}_{3/2}|F^{\prime}=4\rangle transition on the efficiency of \Lambda -GMC.