Generation of optical skyrmionic beams, topological textures of polarization Stokes vector in the transverse plane and propagating at the longitudinal axis provide a versatile platform for topologically nontrivial optical informatics and light–matter interactions. However, their acceleration along curved trajectories is to be studied. In this study, we experimentally demonstrate the first accelerating skyrmion lattices conveyed by Airy structured light, characterized by topologically stable skyrmion textures with self-acceleration along parabolic trajectories. We quantify topological stability using the Skyrme number Nsk, where |Nsk| = 1 represents an ideal skyrmion. Our experiments demonstrate that the skyrmion unit cell can maintain a Skyrme number |Nsk| > 0.9 within a propagation range of ±1.22 zR (zR is the Rayleigh length) upon parabolic acceleration. Notably, the meron structure remains |Nsk| stable within 0.5 ± 0.02 over a significantly extended range of ±3.06 zR. Our work provides a new potential carrier for topologically robust information distribution, particle sorting, and manipulation.
I. INTRODUCTION
The concept of topological solitons, as stable particle-like field configurations, traces its roots to Lord Kelvin’s 19th-century hypothesis of knotted vortices.1–3 This idea was revived in 1961 by Tony Skyrme, who proposed a topological field theory where atomic nuclei emerge as 3D continuous-field solitons—later termed skyrmions—characterized by an integer Skyrme number quantifying their mapping from real space to the three-sphere.1,4,5 Initially studied in high-energy physics,6 skyrmions gained prominence in optics through their analogy to structured light fields with nontrivial polarization topologies.7–9 Especially, it becomes a mainstream to extend more complex topological structures controlled in higher dimensions of free space,10 such as conformal skyrmions, meron lattices,11,12 bimerons,13,14 multiskyrmions and multimerons,15,16 and hopfions,1,17,18 with more flexible control and resilient propagation.
Unlike the condensed-matter-based skyrmions localized in materials for potential data storage or local data processing,6–9 the optical skyrmions possess the ability to propagate in free space, which is invaluable for practical information transfer applications and nontrivial light–matter interactions.19–22 Recently, exploiting advanced technologies of structured light modulation,23–25 both individual skyrmions and skyrmion lattices can be not only generated26–29 but also propagated in free space with controlled topologies.30–35 However, all the experimentally observed free-space skyrmions can only propagate straightly. Controlling topological skyrmions that spatially accelerate along curved trajectories is crucial for higher-level information sorting or transfer but remains a challenge.
Hence, the primary goals of this study are (1) to develop and experimentally validate a method for generating accelerating skyrmions and (2) to quantitatively assess their topological stability through the evolution of Nsk during propagation. It is important to note that these optical skyrmions (or baby skyrmions)36 represent a reduced-dimensional special case of the 3D particle-like skyrmionic hopfion,1 consisting of 2D topological textures in Stokes space. They are characterized by a 3D Stokes vector field . These textures exist on the transverse plane of the light field and are constructed via stereographic projection of a 3D Poincaré sphere, exhibiting continuous radial polarization rotation from the core outward [Fig. 1(a)]. To achieve such structures, we generate a vector beam, specifically a Poincaré beam, through the non-separable superposition of orthogonally polarized Laguerre–Gaussian (LG) beams.30,37,38 Subsequently, accelerating skyrmion lattices are obtained via Fourier transform after cubic phase modulation. Here, the term “skyrmion lattice” refers to a periodic lattice-like array of skyrmion unit cells embedded within a single structured beam, distinct from both (i) an ensemble of independent beams and (ii) ideal lattices with perfect periodicity.11,29,39 Through theoretical simulations and experimental verification, we show that the periodic Gaussian- and vortex-like intensity profiles of Airy and vortex Airy-like beams naturally form periodic skyrmion structures. Notably, these skyrmion lattices maintain robust topological properties within the Rayleigh length of the Airy beams during free-space propagation, confirming their stability. These results not only expand the fundamental understanding of skyrmion dynamics but also open new avenues for practical applications in topological information distribution, particle sorting, and manipulation in nontrivial light–matter interactions.
Concept of periodically accelerating skyrmion lattices. (a) Schematic of periodic skyrmion lattice generation via cubic phase modulation. The original Bloch-type skyrmion texture (Nsk = −1) transforms into square skyrmion lattices, each unit cell retaining Nsk = −1. The color map represents the Stokes vector orientation mapped to the HSL color space, with maximum saturation. Hue encodes transverse orientation ψ = arg(S1 + iS2), and lightness corresponds to S3. (b) Airy and vortex Airy-like beams, exhibiting periodic intensity lattices that are complementary to each other (highlighted in red and blue regions), accelerate along curved trajectories.
Concept of periodically accelerating skyrmion lattices. (a) Schematic of periodic skyrmion lattice generation via cubic phase modulation. The original Bloch-type skyrmion texture (Nsk = −1) transforms into square skyrmion lattices, each unit cell retaining Nsk = −1. The color map represents the Stokes vector orientation mapped to the HSL color space, with maximum saturation. Hue encodes transverse orientation ψ = arg(S1 + iS2), and lightness corresponds to S3. (b) Airy and vortex Airy-like beams, exhibiting periodic intensity lattices that are complementary to each other (highlighted in red and blue regions), accelerate along curved trajectories.
II. CONCEPT AND PRINCIPLE
Returning to the main focus, a natural route to construct accelerating skyrmion lattices leverages the unique properties of Airy beams, which exhibit non-diffracting, self-accelerating intensity distributions with periodic lattice structures in the transverse plane.44,45 While ideal Airy beams require infinite energy, only finite-energy Airy beams can be realized through the Fourier transform of a Gaussian beam modulated by a cubic phase. Introducing vortex phases during this process results in so-called vortex Airy-like beams. Although it is accepted to call such beams Airy as they retain self-accelerating properties due to their cubic phase,46–48 it should be noted that they are actual spiral-phase-contrast Airy beams. As a result, periodic vortex lattices emerge, while the exact Airy function is no longer preserved. Thereafter, the combination of these beams results in the formation of periodic skyrmion lattices (see the supplementary material, S2, for more details).
For ideal Airy beams, the topologically stable skyrmion textures self-accelerate along parabolic trajectories. However, for the finite-energy Airy beams used in this study, topological stability is compromised. As shown in Fig. 1(b), while both simulated Airy and vortex Airy-like beams accelerate along parabolic trajectories, transferring their self-accelerating behavior to skyrmion lattices, significant deformations in the periodic intensity distributions (particularly for vortex Airy-like) occur during propagation. As a result, on the Fourier plane zf, the perfect skyrmion lattices, which exist in the periodic intensity cells of Airy beams as shown in the inset, will inevitably decay during propagation due to these intensity deformations (see the supplementary material, S2, for more details).30
In this context, we further consider the meron structure, as depicted in the inset of Fig. 1(a). This structure, located within the red ring line (the equatorial projection of the Poincaré sphere) of the skyrmion, is characterized by its half-wrapping of the Poincaré sphere. The Skyrme number for the meron, , is not a standard integer topological invariant but rather a numerical index that describes the degree of coverage of the Poincaré sphere. Of particular interest is the potentially enhanced structural stability of meron structures upon propagation, which is localized in skyrmion core regions compared to decaying outer skyrmion domains.
III. EXPERIMENTAL SETUP
To further demonstrate the accelerating properties of skyrmion lattices, the experimental setup is depicted in Fig. 2. A self-locking polarization Mach–Zehnder interferometer, incorporating two beam displacers (BD-1 and BD-2), forms the core of the system. A continuous 532 nm TEM00 laser beam (CNI Laser MSL-III-532) passes through BD-1, splitting into two orthogonally polarized beams directed to separate regions of a spatial light modulator (SLM, HOLOEYE ERIS-NIR-153). These beams are individually modulated using complex amplitude modulation (CAM) holograms corresponding to the Fourier transforms of the desired Airy and vortex Airy-like beams.50 The beam radius w0 and scaling factor b are set as 1 mm and , respectively. A half-wave plate (HWP2, fixed at 45°) ensures horizontal polarization for both beams before modulation by the SLM. After modulation, BD-2 recombines the beams. A quarter-wave plate (QWP1, fixed at 45°) and a Fourier lens (L) with a focal length of 200 mm are used to reconstruct the Airy vector beam at the Fourier plane. The propagation tomographies of the Airy beams are recorded via a digital propagation technique integrated into the CAM holograms, eliminating the need for physical translation of the CMOS camera (Allied Vision Alvium 1800 U-240m) along the z axis.51,52 Furthermore, the camera, combined with polarizers (QWP2, HWP3, and PBS), performs spatial Stokes tomography, capturing the structures of skyrmion lattices in Stokes space.
Experimental setup. The key components include beam displacer prisms (BD), a spatial light modulator (SLM), a camera (CMOS), a half-wave plate (HWP), and a quarter-wave plate (QWP). The upper-right corner inset displays the holograms used in the experiment. The holograms within the dashed box are used to generate the Airy beams, while the hologram outside the box is coherently superimposed with those inside to perform digital propagation measurements. The lower-right corner inset shows the polarization-dependent intensity profiles obtained via spatial Stokes tomography at the Fourier plane.
Experimental setup. The key components include beam displacer prisms (BD), a spatial light modulator (SLM), a camera (CMOS), a half-wave plate (HWP), and a quarter-wave plate (QWP). The upper-right corner inset displays the holograms used in the experiment. The holograms within the dashed box are used to generate the Airy beams, while the hologram outside the box is coherently superimposed with those inside to perform digital propagation measurements. The lower-right corner inset shows the polarization-dependent intensity profiles obtained via spatial Stokes tomography at the Fourier plane.
IV. RESULTS AND DISCUSSION
Figure 3(a) shows the observed 3D profiles of vortex Airy-like (RCP) and Airy (LCP) beams (reconstructed from 26 slices) with a Rayleigh length of zR = 32.701 mm, where both components propagate along diagonal curved trajectories (see the supplementary material, S2, for more details). Transverse intensity profiles below the 3D profiles reveal periodic lattice structures that degrade with propagation, more prominently for vortex Airy-like beams. At z = 100 mm, the vortex lattices even become indistinguishable. Unlike skyrmion beams that propagate straight and form skyrmion tubes along the propagation direction, the skyrmion lattices in this study extend as parallel and curved skyrmion tubes in 3D space, forming curved skyrmion density tubes, as shown in Fig. 3(b). Hence, although 3D polarization distributions have been measured, no 3D topological features like hopfions exist in this study due to the presence of parallel skyrmion tubes.
Experimental results for periodically accelerating skyrmion lattices. (a) 3D intensity profiles of vortex Airy-like (RCP) and Airy (LCP) beams, reconstructed from experimental measurements, with selectively sampled transverse intensity lattices at different propagation distances (0, 32, 64, and 100 mm). (b) 3D skyrmion density distributions corresponding to the self-accelerating intensity lattices. (c) Detailed distributions of the Stokes vector and skyrmion density at different z planes. As the propagation distance increases, skyrmion textures degrade, and positive ρsk values (red) gradually emerge alongside the initially negative ρsk lattices (blue).
Experimental results for periodically accelerating skyrmion lattices. (a) 3D intensity profiles of vortex Airy-like (RCP) and Airy (LCP) beams, reconstructed from experimental measurements, with selectively sampled transverse intensity lattices at different propagation distances (0, 32, 64, and 100 mm). (b) 3D skyrmion density distributions corresponding to the self-accelerating intensity lattices. (c) Detailed distributions of the Stokes vector and skyrmion density at different z planes. As the propagation distance increases, skyrmion textures degrade, and positive ρsk values (red) gradually emerge alongside the initially negative ρsk lattices (blue).
To further evaluate topological stability, Fig. 3(c) illustrates four slices of skyrmion lattices and their corresponding skyrmion densities, focusing on the same three-by-three grid area as in Fig. 1, at different z positions. At the Fourier plane (z = 0), perfectly periodic Stokes vector and skyrmion density distributions are observed, consistent with the theoretical results in Fig. 1(a). However, as propagation distance increases, vortex lattice degradation leads to skyrmion cell deformation, a reduction in the S3 = 1 area [white region of S in Fig. 3(c)], and the emergence of abnormal positive ρsk values (theoretical references for these results are provided in Fig. S3). These results provide an intuitive understanding that the topological quality of the skyrmion lattices gradually degrades during propagation.
To further quantitatively analyze the stability, the Skyrme numbers of three cells of skyrmion lattices (labeled 1, 2, and 3) were calculated, as shown in Fig. 4(a). As the lattices degrade, the Skyrme number Nsk gradually decreases with distance, with cell 2 showing the fastest decline due to its smaller integration area, which becomes insufficient for maintaining accuracy with increasing lattice complexity. Despite these challenges, skyrmion lattices maintain stable acceleration over a range of ±40 mm (±1.22 zR), with . Beyond this range, Nsk drops to approximately −0.6.
Quantitative analysis of accelerating skyrmion and meron lattices. (a) Skyrme number for skyrmion lattices, showing a gradual decrease with propagation distance. (b) Skyrme number for meron lattices, remaining stable over propagation distance. Point data correspond to experimental observations, and solid lines represent theoretical results.
Quantitative analysis of accelerating skyrmion and meron lattices. (a) Skyrme number for skyrmion lattices, showing a gradual decrease with propagation distance. (b) Skyrme number for meron lattices, remaining stable over propagation distance. Point data correspond to experimental observations, and solid lines represent theoretical results.
Interestingly, meron lattices, which occupy the core regions of skyrmion structures, demonstrate a greater level of robustness. As shown in Fig. 4(b), meron lattices maintain stability across the entire measured range (±3.06 zR), with Nsk ≈ −0.5. This stability is achieved at the expense of reduced S3 = 1 regions, as the inner topological structures are better preserved.
Overall, Skyrmion lattices in this work not only inherit the stable acceleration and self-healing properties of Airy beams53–55 (see the supplementary material, S3, for details of self-healing) but also introduce an additional degree of freedom due to their intricate topological structures. These features not only unlock new opportunities for exploring their dynamic behavior but also show promise for practical applications.
V. CONCLUSION
We have experimentally demonstrated a versatile family of optical skyrmion lattices capable of accelerating along curved trajectories in free space. These lattices are generated by imprinting cubic phase modulation onto traditional skyrmion beams. The transverse acceleration rate of these trajectories can be flexibly controlled by adjusting the scaling factor b. Importantly, only LG beams that form skyrmion structures with odd topological charge differences can generate skyrmion lattices.
Leveraging the non-diffracting and self-accelerating nature of Airy beams, skyrmion lattices exhibit stable transverse acceleration and robust topological properties over a propagation range of ±1.22 zR, even under propagation-induced lattice deformation. Greater robustness is observed for Meron lattices, which occupy the core regions of skyrmion structures and remain stable over an extended range of ±3.06 zR.
These unique features establish skyrmion lattices as a promising platform for exploring advanced applications such as particle sorting,56–58 advanced microscopy,57–62 plasma channel induction,57,63 electron beam manipulation,57,58 and guiding electric discharges,64 thereby opening new frontiers in structured light research.
SUPPLEMENTARY MATERIAL
See the supplementary material for more details of Laguerre–Gaussian modes, Airy, and vortex Airy beams.
ACKNOWLEDGMENTS
Singapore Ministry of Education (MOE) AcRF Tier 1 (Grant No. RG157/23), MoE AcRF Tier 1 Thematic (Grant No. RT11/23), Imperial-Nanyang Technological University Collaboration Fund (Grant No. INCF-2024-007), Singapore Agency for Science, Technology and Research (A*STAR) MTC Individual Research Grants (Grant No. M24N7c0080), Nanyang Assistant Professorship Start Up Grant, and National Natural Science Foundation of China (Grant Nos. 12474324, 62075050, and 11934013).
AUTHOR DECLARATIONS
Conflict of Interest
The authors have no conflicts to disclose.
Author Contributions
Haijun Wu: Conceptualization (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Resources (equal); Software (equal); Supervision (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Weijie Zhou: Funding acquisition (equal); Investigation (equal); Methodology (equal); Resources (equal); Software (equal); Supervision (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Zhihan Zhu: Funding acquisition (equal); Investigation (equal); Methodology (equal); Resources (equal); Software (equal); Supervision (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Yijie Shen: Conceptualization (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Resources (equal); Software (equal); Supervision (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal).
DATA AVAILABILITY
The data that support the findings of this study are available from the corresponding author upon reasonable request.