Direct Generation of an Array with 78400 Optical Tweezers Using a Single Metasurface

Abstract: Scalability remains a major challenge in building practical fault-tolerant quantum computers. Currently, the largest number of qubits achieved across leading quantum platforms ranges from hundreds to thousands. In atom arrays, scalability is primarily constrained by the capacity to generate large numbers of optical tweezers, and conventional techniques using acousto-optic deflectors or spatial light modulators struggle to produce arrays much beyond $\sim 10,000$ tweezers. Moreover, these methods require additional microscope objectives to focus the light into micrometer-sized spots, which further complicates system integration and scalability. Here, we demonstrate the experimental generation of an optical tweezer array containing $280\times 280$ spots using a metasurface, nearly an order of magnitude more than most existing systems. The metasurface leverages a large number of subwavelength phase-control pixels to engineer the wavefront of the incident light, enabling both large-scale tweezer generation and direct focusing into micron-scale spots without the need for a microscope. This result shifts the scalability bottleneck for atom arrays from the tweezer generation hardware to the available laser power. Furthermore, the array shows excellent intensity uniformity exceeding $90\%$, making it suitable for homogeneous single-atom loading and paving the way for trapping arrays of more than $10,000$ atoms in the near future.

• The paper presents a metasurface method that directly generates a 280x280 (78,400) optical tweezer array with diffraction-limited focus and high light efficiency.
• It employs nanostructured phase control using CMOS-compatible fabrication and advanced simulations to achieve >90% intensity uniformity and 58.8% light utilization.
• The work significantly advances scalable quantum processing by shifting hardware bottlenecks from optical components to available laser power.

Direct Generation of an Array with 78,400 Optical Tweezers Using a Single Metasurface

Introduction

Recent advances in neutral atom platforms for quantum information processing have highlighted the critical bottleneck posed by scalable optical tweezer generation. Conventionally, acousto-optic deflectors (AODs) and spatial light modulators (SLMs) have enabled the creation of atom arrays containing up to $\sim 10,000$ tweezers, but these technologies require additional bulk optics for focusing and are constrained by pixel number and diffraction efficiency. Metasurface photonics, leveraging subwavelength phase control, provides a single, planar solution capable of arbitrarily shaping the wavefront and directly focusing light into micron-scale optical tweezers. This paper details the experimental realization of a $280 \times 280$ ($78,400$) array of diffraction-limited optical tweezers based on a single metasurface, demonstrating superior scalability and light utilization efficiency compared to AODs/SLMs and shifting the bottleneck for quantum processor scaling from tweezing hardware to available laser power (2512.08222).

Figure 1: Optical tweezer generation approaches: (a) AODs, (b) SLMs, and (c) metasurfaces with direct lensless focusing via nanostructured phase control.

Metasurface Design and Meta-Atom Engineering

The metasurface is fabricated on a $10~\text{mm} \times 10~\text{mm}$ fused silica substrate, featuring a $5~\text{mm}$ diameter active area. Each meta-atom is a square, silicon-doped silicon nitride nanopillar ($H = 1150$ nm, lattice constant $P = 430$ nm). Polarization independence is ensured by the cross-sectional geometry, while the refractive index ($n = 2.3$) is optimized for near-infrared (852 nm) operation, relevant for $^{87}$Rb trapping. Phase modulation from $-\pi$ to $\pi$ with $>85\%$ transmittance is verified using FDTD simulations, underpinning precise amplitude/phase control across the metasurface. The phase–size response determines the meta-atom distribution for spatial wavefront engineering.

Figure 2: FDTD-calculated phase shift $\phi$ and transmittance $T$ versus meta-atom size $L$, enabling high-efficiency, full-range phase coverage.

The global phase profile for the $280 \times 280$ tweezer array is computed via the Weighted Gerchberg-Saxton algorithm, with iterated forward/backward propagation using the angular spectrum method. Convergence yields a phase mask sampled on $11,628 \times 11,628$ grid, requiring blockwise FFT due to large matrix size. Simulation results predict an Airy disk radius of $0.9~\mu$m and array intensity nonuniformity of $2.45\%$.

Fabrication and Optical Characterization

Fabrication utilizes CMOS-compatible e-beam lithography and plasma etching, yielding high aspect ratio ($>11:1$), minimal roughness nanopillars. The laser characterization setup directs a collimated $852$ nm beam ($7.2$ mm waist) onto the metasurface, which focuses the light at $3.49$ mm to produce the tweezer grid. The focal plane is mapped via a microscope objective (NA = 0.65, FoV = 0.45 mm) and a high-resolution CCD.

Figure 3: (a) SEM image of metasurface nanopillars. (b) Subarea of $280 \times 280$ tweezer array showing uniformity and periodicity. (c) Optical image of 5 mm-diameter metasurface on substrate.

Spot spacing is $4.258 \pm 0.044~\mu$m (design target: $4.3~\mu$m). Light utilization is quantified—67.5% of incident photons are routed into the first diffraction order (cf. SLM at 45%), and, after accounting for background and non-illuminated regions, the net efficiency is 58.8%. The intensity uniformity across the array has a measured standard deviation of 9.4%, in agreement with robust large-scale fabrication.

Array Performance: Uniformity, Diffraction-Limited Focus, and Scaling Limits

Optical quality metrics are analyzed for each individual tweezer via point spread function (PSF) fitting. Strehl ratios exceeding 0.8 are achieved for 99.7% of spots. The Airy disk radius is $1.017 \pm 0.038~\mu$m, corresponding to a NA of 0.51. The dominant source of deviation from simulations is finite incident beam size.

Figure 4: (a) Normalized intensity profile across array. (b) Histogram of intensity uniformity (9.4% stdev). (c) Histogram of Airy disk waist approaching diffraction limit.

Compared to AOD/SLM platforms, metasurfaces enable a tightly focused, homogeneously intense, and vastly larger number of tweezers in a compact format without external microscope objectives. The demonstrated setup shifts the limiting factor to available laser power. For $^{87}$Rb tweezers, $0.8$ mW/spot is required for high-loading efficiency. Commercial $850$ nm lasers offer $50$ W; after losses, $\sim 33,000$ tweezers can be simultaneously loaded at full depth. The metasurface exceeds this number by more than a factor of two, affirming hardware scalability.

Comparative Analysis and Implications

The present demonstration joins and exceeds prior metasurface atom array implementations. Previous works realized $5 \times 5$ and $3 \times 3$ tweezer arrays, with the largest prior effort reporting $\sim$360,000 tweezers (but with lower pixel-per-tweezer and metasurface area constraints) [holman2025]. Compared to SLMs/AODs, which are limited by pixel count, power efficiency, and the need for high-NA bulk optics, metasurfaces offer integration, compactness, polarization-independence, and highly parallelizable fabrication. By adopting a higher pixel-per-tweezer ratio, the work optimizes beam quality and ensures practical deployability outside of vacuum environments, where minimum working distances are geometric constraints.

The implications for quantum simulation and computing are substantial: metasurface-based tweezer arrays now exceed the hardware scale required for effective quantum error correction in logical qubit encoding schemes [bluvstein2025architectural]. Homogeneous intensity and near-diffraction-limited focusing are compatible with high-fidelity quantum gate operation, as demonstrated in large neutral atom systems [evered2023high]. As optical power scaling continues, metasurfaces enable further increase in atom number, paving the way for universal, fault-tolerant quantum computing on atom arrays.

Conclusion

This paper demonstrates direct generation of a $280 \times 280$ optical tweezer array using a single metasurface, with $>90\%$ intensity uniformity, 58.8% light utilization efficiency, and diffraction-limited spot size. The technology removes the microscope bottleneck and fundamentally shifts scalability constraints to optical power availability. The design and fabrication approaches illustrated are readily extensible; with continued improvement in high-power laser systems and metasurface engineering, practical 10,000+ qubit atom array quantum processors are achievable, advancing both fundamental quantum science and engineered quantum computation.