Local polarization control of optical tweezer arrays
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.
Kai: I'm Kai, and with me are Mira and Lev, guest researcher.
Mira: Today's paper: "Local polarization control of optical tweezer arrays".
Kai: The gist: This work demonstrates independent, site-resolved linear polarization rotation across an optical tweezer array by exploiting the local birefringence tunability of a spatial light modulator,
Mira: First, who's behind it and why it matters.
Title and authors: Kai: So we're looking at the paper "Local polarization control of optical tweezer arrays," and the authors are Kai Sun, Richard Bing-Shiun Tsai, Andrew Winnicki, Yuan Le, Kon H. Leung, Nelson Darkwah Oppong and Manuel Endres.
Mira: The title itself is very direct about what they’re tackling: local polarization control of optical tweezer arrays. It immediately signals that they are focused on getting individual control over the polarization of each trap light source.
Kai: And it sets the stage because before this, controlling polarization for these things was considered a major experimental challenge, so this work is showing how to overcome that hurdle using an SLM.
Lev: From an error correction standpoint, having site-resolved control over the addressing beams means you can tailor your qubit operations precisely to the local environment of each atom.
Mira: It’s about moving beyond just using a fixed laser setup and instead having a flexible addressing scheme that lets you program polarization patterns onto the array.
Kai: Essentially, they are demonstrating how to use the local birefringence of an SLM at an intermediate image plane to achieve this independent rotation for every tweezer.
Lev: That's cool because it means you can potentially create tailored Hamiltonians for different regions of your array just by changing those SLM settings.
Mira: Yes, they show two main ways to program this: one uniform pattern that hits a magic-angle condition everywhere, and another staggered pattern to isolate groups of atoms spectrally.
Kai: That staggered pattern is interesting because it lets you create distinct spectral zones within the array, which leads directly into their idea for zone architecture later on.
Lev: If you can spectrally isolate groups of atoms using polarization patterns, that’s a powerful tool for defining boundaries in your quantum processor design.
Mira: So the implication here is that we can start treating polarization not just as a passive property, but as an active control parameter for the trapped system.
The paper's summary: Kai: This section of the paper summarizes how they set up this experiment using their global eight hundred thirteen-nm laser beam diffracted by an AOD to generate a one-dimensional array of optical tweezers <ref:2610.12406#pg1>.
Mira: They then place an SLM at an intermediate image plane, combined with a pair of QWPs oriented at forty-five degrees to the SLM’s principal axes, which allows for arbitrary linear polarization rotations of each tweezer <ref:2610.12406#pg2,at an intermediate image plane, combined with a pair of QWPs oriented>.
Kai: The rotation itself is directly related to half the phase phi imprinted by the illuminated SLM pixels on those traps.
Lev: So when you look at their setup, it’s a standard optical tweezer array relay system, but with this added layer of programmable polarization control introduced at that intermediate plane.
Mira: And they do calibration by measuring the differential light shifts on the narrow intercombination transition, 1S0 3P1, to extract theta, which is the angle between the linear tweezer polarization and a bias magnetic field B <ref:2610.12406#pg3,the angle between the linear tweezer polarization>.
Kai: The main result here is that they successfully program these individual tweezer polarizations in two configurations: a uniform pattern setting the magic-angle condition, and a staggered pattern for spectral isolation.
Lev: So they’ve shown the mechanism works on this specific atomic transition, which is good because it’s a clean spectroscopic probe for what they are trying to engineer.
Mira: The summary really boils down to: we use an SLM to program polarization, measure the resulting light shift differences, and then apply corrections based on that measurement.
The paper's improvements: Kai: The paper suggests a few ways to improve this, starting with how they handle cooling. They focus on engineering magic-angle conditions with local polarization tuning for sideband cooling.
Mira: Under these conditions, the differential light shifts vanish, or nu = zero which decouples the sensitivity of the cooling resonance from site-to-site tweezer intensity non-uniformity <ref:2610.12406#pg3>.
Lev: That decoupling is huge; it means you can achieve uniform loading much more efficiently because you don't have to worry about every single trap being perfectly tuned to the same frequency.
Kai: They find that when these local polarization corrections are applied, the optimal loading condition collapses to a single global cooling laser frequency across all traps.
Mira: They pinpoint this optimization occurs at theta = theta M, which is forty-nine point two degrees, and that's where the differential trap light shifts are nulled <ref:2610.12406#pg2>.
Lev: So they’ve found the specific angle that makes the cooling mechanism work uniformly across the whole array without having to compensate for every site individually.
Kai: Then there’s their idea for zone architecture, where atoms are cooled in one zone and then transported to a second zone where polarization rotation spectrally detunes the transition away from resonance with the cooling laser.
Mira: This isolation strategy minimizes crosstalk between zones because the atoms in the computing zone aren't being hit by the cooling light at all.
Lev: That way, you can perform different types of operations sequentially without worrying about one part interfering with another.
Kai: And for transport, they show that coherent transport works up to theta = pi/eight but as the change increases to pi/two survival drops sharply because of spatial mode deformation at the interface.
Mira: They suggest using a linear phase ramp between superpixels to smooth out that polarization transition over a finite distance, which recovers the zero baseline for ramp lengths greater than two point zero times the wavelength <ref:2610.12406#pg1>.
Lev: That ramp is a practical way to manage the interface; it trades perfect sharp transitions for controlled, manageable ones during transport.
Conclusion: Kai: So to wrap up on "Local polarization control of optical tweezer arrays," they establish local polarization control for tweezer arrays as a new tool for quantum science applications, which is really important.
Mira: They’ve shown that this capability establishes tweezer polarization as a programmable degree of freedom alongside position and intensity, which is something we can actively program.
Lev: This capability is expected to find use where polarization shapes the trapping potential or the strength of atom-light coupling itself, which opens up new avenues for control.
Kai: It sets the stage for further exploration of quantum metrology and quantum simulation with dynamically reconfigurable strontium arrays in two dimensions, which is a big area.
Mira: The main thing is that we have a programmable per-site degree of freedom that's not just position or intensity, making it much richer.
Lev: We’ll need to see how this translates into practical error correction schemes for these dynamically reconfigurable systems as they try to build them.
Kai: That’s what we're looking at next time, moving from the control mechanism to actually putting these zones together in a functional processor.
California Institute of Technology · Oratomic
physics.atom-ph, physics.optics, quant-ph
Submitted: 2026-10-08
Updated: 2026-10-08
Comments: 11 pages, 5 figures
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
Importance score: 92/100
The gist: The gist: This work demonstrates independent, site-resolved linear polarization rotation across an optical tweezer array by exploiting the local birefringence tunability of a spatial light modulator,
Key concepts
- Liquid-Crystal Spatial Light Modulator (SLM)
- The SLM acts as a high-resolution birefringent phase retarder placed at an intermediate image plane. It is used to program specific polarization patterns onto the optical tweezers, allowing researchers to control the linear polarization of each individual tweezer site independently. This enables precise local manipulation of light properties.
- Magic-Angle Sideband Cooling
- This cooling technique uses local polarization tuning to engineer a 'magic-angle condition' where state-dependent differential light shifts vanish (ν=0). This decouples the cooling resonance from intensity variations across the trap array, allowing for uniform sideband cooling even when trap intensities are non-uniform.
- Zone Architecture
- The study proposes a two-zone 2D architecture. In one zone, atoms are cooled to the motional ground state using specific polarization settings. In a second zone, polarization is rotated to spectrally detune the transition from the cooling laser, isolating atoms and enabling coherent transport between zones for quantum processing.
Terminology
Summary
The gist: This work demonstrates independent, site-resolved linear polarization rotation across an optical tweezer array by exploiting the local birefringence tunability of a spatial light modulator, enabling magic-angle sideband cooling and dynamic transport between distinct polarization zones.
Local Polarization Control
The researchers demonstrate independently programmable polarization rotation at the individual sites of an optical tweezer array using a liquid-crystal spatial light modulator (SLM) at an intermediate image plane The SLM functions as a birefringent phase retarder with high spatial resolution, enabling us to control the polarization of each tweezer individually They program polarization patterns onto the tweezer array in two configurations: (i) a uniform polarization pattern that sets every site to the tensor magic-angle condition for the 1S0 ↔ 3P1 intercombination transition, and (ii) a staggered pattern in which two groups of atoms are spectrally isolated
Experimental Setup and Calibration
The experimental setup involves a global 813-nm laser beam diffracted by an acousto-optic deflector (AOD) to generate a one-dimensional array of optical tweezers An SLM placed at an intermediate image plane, combined with a pair of QWPs, allows for arbitrary linear polarization rotations of each tweezer The rotation is half the phase φ imprinted by the illuminated SLM pixels To calibrate the relationship between experimentally inferred θ and the applied voltage on the liquid crystal pixels, they perform spectroscopy on the narrow intercombination transition (1S0 ↔ 3P1) to observe differential light shifts The fit assumes that the atom experiences linearly polarized light
Sideband Cooling and Uniformization
Local polarization tuning enables sideband cooling by engineering magic-angle conditions with local polarization tuning Under this condition, the state-dependent differential light shifts vanish (ν = 0), decoupling the sensitivity of the cooling resonance from site-to-site tweezer intensity non-uniformity The optimal loading condition collapses to a single cooling laser frequency across all traps when local polarization corrections are applied They find that differential trap light shifts are nulled at θ = θM = 49.2(4)◦
Zone Architecture and Transport
The study proposes a zone-based 2D architecture for a strontium quantum processor leveraging local polarization programmability In the cooling zone, atoms are trapped in AOD-generated tweezers tuned to θ ≈ θM and sideband cooled to the motional ground state Outside of the cooling zone, polarization rotation spectrally detunes the 1S0 ↔ 3P1 transition away from resonance with the cooling laser, isolating those atoms from the cooling light and minimizing crosstalk They demonstrate coherent transport between distinct polarization zones For atom transport across polarization zones, they observe no measurable degradation in survival for polarization changes up to ∆θ = π/8, but atom survival drops markedly when ∆θ increases from π/8 to π/2 A linear phase ramp between two superpixels is introduced to smooth the polarization transition over a finite distance, recovering the ∆θ = 0 baseline for ramp lengths ≳ 2.0w
Polarization Noise and Stabilization
They characterize polarization noise using optical polarimetry and atomic spectroscopy, finding mrad-level polarization noise and drifts that closed-loop atomic feedback further suppresses on long timescales Active stabilization via closed-loop feedback updates the superpixel phases to steer the probed line centers of the 1S0 ↔ 3P1 carrier transition to the free-space value, achieving a standard deviation of 20 kHz with feedback This polarization-angle noise translates to effective frequency noise via the differential light shift slope ∂ν/∂θ, resulting in a standard deviation of 3.0(4) mrad at the nominal trap depth
Conclusion and Outlook
The work establishes local polarization control for tweezer arrays as a new tool for quantum science applications Their findings set the stage for further exploration of quantum metrology and quantum simulation with dynamically reconfigurable strontium arrays in 2D The results establish tweezer polarization as a programmable, per-site degree of freedom, alongside position and intensity This capability is expected to find use where polarization shapes the trapping potential or the strength of atom–light coupling
How it works
The method exploits a liquid-crystal spatial light modulator (SLM) at an intermediate image plane to control the polarization of each tweezer individually The SLM functions as a birefringent phase retarder with high spatial resolution, enabling us to control the polarization of each tweezer individually This is achieved by programming polarization patterns onto the tweezer array in two configurations: (i) a uniform polarization pattern that sets every site to the tensor magic-angle condition for the 1S0 ↔ 3P1 intercombination transition, and (ii) a staggered pattern in which two groups of atoms are spectrally isolated
Sideband Cooling Mechanism
Sideband cooling is achieved by engineering magic-angle conditions with local polarization tuning The researchers employ three independent beams aligned to the tweezers, where the cooling beams possess the appropriate polarization component to drive π-transitions The optimization of the cooling parameters involves iteratively adjusting tweezer polarizations to achieve the best cooling performance for each beam
Atom Transport Strategy
Atom transport between zones is investigated by programming a bipartite profile featuring a sharp polarization change of ∆θ between adjacent zones When the polarization change increases from π/8 to π/2, atom survival drops markedly due to spatial mode deformation at the sharp interface A linear phase ramp is introduced between two superpixels to smooth the polarization transition over a finite distance, which recovers the ∆θ = 0 baseline for ramp lengths ≳ 2.0w
Noise Suppression
Polarization noise is characterized using optical polarimetry and atomic spectroscopy, finding mrad-level polarization noise and drifts that closed-loop atomic feedback further suppresses on long timescales The feedback protocol uses the prior calibration of ν and θ versus SLM pixel voltage for each tweezer site to null any accumulated differential light shift Over a 16-hour duration, the stabilized tweezers show both a smaller site-to-site spread in line center and a two-fold reduction in temporal fluctuation
Zone Architecture Implementation
The proposed zone architecture envisions atoms being trapped in AOD-generated tweezers tuned to θ ≈ θM and sideband cooled to the motional ground state in the cooling zone Atoms are then transported by the same tweezers to a computing zone for quantum information processing Polarization rotation spectrally detunes the 1S0 ↔ 3P1 transition away from resonance with the cooling laser, isolating those atoms from the cooling light and minimizing crosstalk
Polarization Noise Characterization
The polarization-angle noise is measured by probing the carrier peak width, which encodes more information than just the peak center position The model for the peak width shows that it is proportional to the trapping laser intensity (trap depth), allowing them to extract an estimate of the polarization fluctuation of 3.0(4) mrad This corresponds to a polarization angle standard deviation of 3.
Improvements for AI systems
-
How to achieve
precise quantum state engineering of trapped particles
by exploitingthe polarization dependence of anisotropic vector and tensor light shifts.
This allows for controlling qubit states based on polarization-sensitive atomic properties in 88Sr arrays. -
Implement
sideband cooling under a magic-angle condition
by using local polarization tuning tohomogenize differential light shifts imparted by the 813-nm tweezers,
enabling efficient cooling even in spatially inhomogeneous arrays. -
Develop a
polarization-defined zone architecture
for quantum processors where atoms are addressed selectively by laser frequency, as demonstrated by creating patterns that facilitatemagic-angle sideband cooling on odd-index tweezers while shifting the resonance of evenindex tweezers 0.4 MHz away.
-
Enable
coherent transport between distinct polarization zones
for trapped atoms, allowing them to survive transport across boundaries where the polarization changes, as shown by observing survival up to∆θ = π/8
and recovering baseline survival at∆θ = π.
-
Implement
closed-loop stabilization referenced to the atomic transition frequency,
reachingmrad-level stability
for tweezer polarization noise, by usingsite-resolved, in-situ spectroscopy of the 1S0 ↔ 3P1 transition
to provide feedback. -
Create a system capable of characterizing and mitigating polarization noise by modeling the peak width dependence on trap depth using the equation
FWHM = aW + p(1 − a)2W squared + cσ squared,
allowing for estimation ofthe polarization fluctuation of 3.0(4) mrad
at nominal trap depth. -
Design an AI system to optimize cooling parameters by iteratively adjusting tweezer polarizations based on the observation that
the axial beam yields the largest temperature reduction
and tuning them to achieve acommon operating frequency near-resonant with the radial red sideband.
Abstract
Optical tweezer arrays trapping ultracold atoms and molecules are a versatile quantum science platform with broad impact in quantum simulation, quantum computation, and quantum metrology. The polarization dependence of anisotropic vector and tensor light shifts offers a degree of freedom for precise quantum state engineering of trapped particles. Similarly, state-selective operations depend on the polarization of the addressing light through the atom--light coupling strength. Yet, methods for controlling the polarization of individual tweezers or local addressing beams remain an experimental challenge. Here, we demonstrate independent, site-resolved linear polarization rotation across an optical tweezer array by exploiting the local birefringence tunability of a spatial light modulator. Applying this capability to an array of 88 atoms, we homogenize differential light shifts imparted by the 813-nm tweezers to the narrow 1 S 0 cubed P 1 transition, enabling sideband cooling under a magic-angle condition. Furthermore, we show dynamic transport of trapped atoms across distinct polarization zones with high survival and preserved coherence. Finally, we characterize tweezer polarization noise and demonstrate closed-loop stabilization referenced to the atomic transition frequency, reaching mrad-level stability. Our work establishes a technique for manipulating polarization-sensitive atomic and molecular transitions in tweezer architectures, with immediate applications to optical tweezer clocks and multi-zone quantum processors.
Sources
- Quantum error correction with the toric code
- A Mid-Infrared Platform Based on Strontium Tweezer Arrays
- Gate-based Readout and Cooling of Neutral Atoms
- Theory of Scalable Spin Squeezing with Disordered Quantum Dipoles
Related papers
- A nuclear clock synchronized to 229 Th
- Dual-Platform Precision Measurement of the 3 2D 5/2 to 4 2S 1/2 g-Factor Ratio in 40 Ca+
- Absolute frequency measurement of a 176 Lu+,(cubed D 1) standard against the NRC-FCs2 fountain with 2.6 times10-16 uncertainty
- Long-lived giant circular Rydberg atoms at room temperature
- Quasiperiodic nondipole ionization dynamics in the x-ray stabilization regime
- The PairInteraction Toolkit for Modeling Rydberg Physics in Alkali and Alkaline-Earth-Like Atoms