betaFeedback

WorksPhysics and Astronomy

Manipulating ferroelectric topological polar structures with twisted light

Abstract

The dynamic control of novel states of matter beyond thermodynamic equilibrium is a fundamental pursuit in condensed matter physics. Intense terahertz fields have enabled metal-insulator transitions, superconductivity, quantum paraelectric ferroelectricity, and room-temperature magnetization via circularly polarized terahertz electric fields. These effects hinge on the excitation of infrared-active soft phonon modes by terahertz fields. Expanding this concept, recent theory suggests that ferroelectric polarization may be manipulated through terahertz twisted light, transferring orbital angular momentum to create ferroelectric skyrmions. Our study experimentally demonstrates that such control is possible in quasi-2D ferroelectric CsBiNb2O7 using twisted UV light with orbital angular momentum (OAM). By resonantly exciting both the ferroelectric mode and the octahedral tilting mode, twisted UV light dynamically modulates the ferroelectric polarization. We employ in-situ X-ray Bragg coherent diffractive imaging, twisted optical Raman spectroscopy, and density functional theory to three-dimensionally resolve ionic displacement fields and polarization texture changes. Our observations reveal deterministic, reversible twisted light-induced strain and ionic displacements within the unit cell, causing substantial microscopic polarization changes. This interaction between twisted photons, phonon modes, and induced ionic displacements breaks symmetry and stabilizes a non-equilibrium ferroelectric phase with topological solitons. These findings offer a new path to control ferroelectricity and magnetism, opening avenues for novel optoelectronic devices such as ultrafast non-volatile memory switches by using light to coherently control ferroic states.

Key points: Twisted light to control polarizations in quasi-two-dimensional FE materials, Bragg coherent X-ray diffractive imaging, data storage, information processing, and energy harvesting

Main

In recent years, there has been a surge in condensed matter research towards nontrivial topologically-protected dipolar textures. These textures have predominantly been found in bulk materials, nanocrystals, and epitaxial heterostructures, where precise electrostatic, magnetostatic, and elastic boundary conditions are enforced on crystalline ferroelectric (FE) and magnetic materials. Since the prediction of flux-closure domain structures in magnetic films in the 1940s [8, 9], parallels have been drawn in predicting and observing real-space topologically-protected dipolar textures such as flux-closures [10], center domains [11], vortices [12, 13], skyrmions [14], merons [15], bubbles [16], and hopfions [17] in FEs.

The typically smaller sizes of these FE domains, compared to their magnetic counterparts, offer promising applications in high-density information storage, thin-film capacitors, actuators, and other electronic devices [9, 8, 15]. Recent theoretical and experimental studies have indicated the presence of Bloch points (BPs) in several polar systems. For instance, Salje and Scott found Bloch lines and BPs within polar domain walls in SrTiO3\mathrm{SrTiO}_{3} [18], while Morozovska et al. [19] observed BPs in BaTiO3\mathrm{BaTiO}_{3} nanoparticles with SrTiO3\mathrm{SrTiO}_{3} shells under certain conditions. Gao et al. [6] demonstrated that an optical vortex beam could generate skyrmion states in ultrathin Pb(Zr0.4Ti0.6)O3\mathrm{Pb}(\mathrm{Zr}_{0.4}\mathrm{Ti}_{0.6})\mathrm{O}_{3} films, leading to the formation of BPs. Despite these advancements, experimentally observing and manipulating topological textures such as vortices, merons and BPs in FE systems remain challenging as they all require strict symmetry constraints and control of the vectorial nature of the applied electric field.

In our earlier work, we demonstrated the volumetric observation of ferroelectric (FE) domains and polar vortices in isolated BaTiO3_3 nanoparticles [12, 20, 21]. Additionally, other studies have experimentally observed phenomena such as negative capacitance and the emergence of chirality in polar vortices and skyrmions, originating from sequences of achiral materials, using resonant soft X-ray diffraction-based circular dichroism and optical measurements [22, 23, 24, 25]. Recent advances, including phase field simulations [26] and aberration-corrected scanning transmission electron microscopy, have further shown that polar BPs can be stabilized in substrate-induced tensile-strained ultrathin FE PbTiO3_3 films.

Despite these significant advancements, the experimental observation and manipulation of polar textures, particularly in a nondestructive manner and in three dimensions, remain challenging endeavors [27, 12]. Furthermore, the ability to apply inhomogeneous electric fields without the use of electrodes, while simultaneously imparting strain inhomogeneities (e.g., tensile, compressive, twist), has proven elusive. Traditionally, the control, stabilization, and manipulation of such FE topological structures have been achieved through substrate-induced strain in ferroelectric/dielectric superlattices [13, 14, 22] or epitaxial thin films [26, 15] under external electric fields. However, these methods face significant challenges, as they are constrained by mechanical boundary conditions rigidly defined by the atomic structure of the substrate or proximity layer. Moreover, manipulating these topological structures often requires the use of AFM contact tips as gates or the design and fabrication of electrodes with high precision to apply nonlinear electric fields at nanoscale resolutions, which is technically demanding.

In this work, we address these challenges by experimentally demonstrating the stabilization of polar BPs due to TL-induced tensile strain in 2D FE flakes using coherent diffractive imaging. We observed the internal 3D structures of these BPs and their reversible transition from BPs to merons and back to BPs. Moreover, we revealed the phenomenon of inducing dynamical steady-state strain inhomogeneities. Our findings not only advance the fundamental understanding of topological textures in FEs but also open new pathways for their application in next-generation ultrafast technologies.

It has been established theoretically [6, 7] that TL in the THz regime can excite and drive dynamical polar skyrmions and transfer its nonzero winding number to FE ultrathin Pb(Zrx_xTi1−x_{1-x})O3_3 films. This interplay between the orbital angular momentum (OAM) [28] of TL and electric dipoles in FE material leads to a distinctive alignment within the polarization fields. While the TL electric field EOAM\mathbf{E}_{\mathrm{OAM}} exhibits uniformity across layers, the zz-component of the polarization exhibits layer-dependent variations, contrasting with the consistent in-plane components. The crux of this interaction is the electrostatic coupling −p⋅EOAM-\mathbf{p}\cdot\mathbf{E}_{\mathrm{OAM}}, which governs the energetically favorable orientation of the dipoles leading to changes in the out-of-plane polarization PzP_z. This subtle alignment, in turn, drives the redistribution of bound charges, tailoring the zz-component of the electric dipoles and hence, the evolution of polar skyrmions. Theoretical predictions suggest this coupling can induce polarization modulations ranging from 0.2 to -1.0 C/m2^2 in response to targeted ionic displacements within the PZT films. Inducing and detecting such ionic motion in an isolated free-standing quasi-two-dimensional FE material is the main idea of the experiment presented here and illustrated in

Figure 1.

Figure 1. Schematic of the manipulation of FE using vortex field from twisted light and its experimental detection. (a) The CsBiNb2_2O7_7 crystal unit responsible for FE, shown in the absence of twisted light. When driven by a plane-polarized 375 nm UV field in a Laguerre-Gaussian (LGℓ_\ell) mode carrying orbital angular momentum (OAM) quantized at ℓℏ\ell\hbar per photon, it induces orbital atomic motions around the beam’s propagation axis. This motion creates a net dipole moment per unit cell and lowers symmetry through resonant excitation of both the zone-center and the zone-boundary octahedral tilting, FE modes. (b, c) Twisted coherent UV beam topological charges ℓ=0\ell= 0 and ℓ≠0\ell\ne0 drive the zone-center (b) and the zone-boundary octahedral tilting (c) modes, respectively. (d) Experimentally observed changes in polarization and atomic displacements along the [2] direction in CBNO arising due to the phonon modes depicted in (b & c).

Layered perovskite such as the Dion-Jacobson (DJ) phase CsBiNb2_2O7_7 (CBNO) are not only a significant class of FEs but also provide an ideal system for exploring the mechanisms that induce polar structures, especially under OAM illumination. These mechanisms include second-order Jahn-Teller-driven cation displacements, cation ordering patterns, and combinations of nonpolar octahedral rotations. CBNO is an improper FE with P21amP2_1am symmetry at room temperature [29]. It is particularly suited for probing electric-field-driven dynamics due to its several zone-center phonon modes within the frequency range accessible to modern terahertz sources. As illustrated in Figure 1(a), theory suggests that the interplay between Γ-point zone-center phonons and M-point zone-boundary phonons in CBNO drives its critical phase transitions [30]. Specifically, the transition from P4/mmmP4/mmm to PmamPmam is associated with the stabilization of an M5_5 zone-boundary phonon, while the transition to P21amP2_1am symmetry requires the cooperative action of a Γ5_5 zone-center FE mode and an M-point zone-boundary mode, accounting for the observed symmetry change. Surface-sensitive experiments [29] and density functional theory (DFT) predict significant FE behavior in the DJ phase of CBNO, with a polarization P≈30−40 μC cm−2\mathbf{P} \approx30-40\,\mu\mathrm{C}\,\mathrm{cm}^{-2}, much higher than that in the n=2n = 2 Aurivillius phase SrBi2_2Ta2_2O9_9 [30]. Structural studies also confirm that the adopted room temperature polar, orthorhombic structure (P21amP2_1am symmetry, 2at×2at×ct\sqrt{2}a_\mathrm{t} \times\sqrt{2}a_\mathrm{t} \times c_\mathrm{t}) is stable up to at least 1000 °C [31]. This polar structure originates from a tetragonal structure through the rotation of NbO6_6 octahedra around both an in-plane axis ([110]t_\mathrm{t}, a−a−c0a^-a^-c^0 in Glazer notation, corresponding to the (M5−_5^-) rotational mode) or the tetragonal axis [1]t_\mathrm{t} (a0a0c+a^0a^0c^+, corresponding to the (M2+_2^+) rotational mode), along with the FE displacement of Bi3+^{3+} cations relative to its neighboring O2−^{2-}- ions along the polar aa-axis [32, 33] as depicted in Figure 1(a). This structure is analogous to FE SrBi2_2Ta2_2O9_9 and related off-stoichiometric phases [34].

Our recent experimental investigations into pyroelectric measurements using both HAADF-STEM and XRD have demonstrated the presence of giant pyroelectricity [35] and room-temperature FE with in-plane domain in free-standing quasi-two-dimensional CBNO flakes [36]. However, experimental operando demonstration of its three-dimensional displacement fields and domain structure due to FE behavior has been elusive until now [32, 33].

Moreover, distinguishing between types and volumetric probing of topological FE polar structures and solitons and 3D induced strain under OAM illumination (Figure 1(b-Figure 1c)) necessitates the development of non-destructive, high spatial resolution probes, namely Bragg coherent diffractive imaging (BCDI) [20, 37] capable of tracking changes in inter-atomic spacing and microscopically induced polarization (Figure 1(d)) in a quasi-2D free-standing FE crystals. These changes in atomic spacing are in line with a previously reported study on temperature-dependent analysis in CBNO [38].

In this article, we present experimental evidence demonstrating the manipulation of FE and tracking d002d_{002} spacing by combining in-situ 375 nm continuum twisted UV light, generated with a tabletop source, with X-ray BCDI [2, 37] (see Figure 2(a) and in Materials and Methods) and optical Raman spectroscopy. We use BCDI to reconstruct 3D atomic displacements and strain maps of CBNO nanoflakes and gain an understanding of the unusual TL control of FE polar structures. Theoretical computations are consistent with these observations and provide additional insights into the origin of improper FE in 2D Dion–Jacobson systems. Our observations show FE domains and Z2×Z4_2 \times Z_4 polar vortex loops spanning the volume of CBNO nanocrystals, suggesting the use of CBNO as an alternative to rare-earth manganites for studies of the Kibble-Zurek mechanism in condensed matter. By varying the topological charge (see Figure 2(c)) of TL, we can manipulate FE and study the dynamics of FE domains, vortex loops, and the vorticity (winding number) of FE polarization near the loops. These loops appear to be stable and locally pinned due to the topologically protected FE vortices. We identify four antiphase FE domains with left/right in-plane polarization configurations that form a vortex or an antivortex. Our results suggest that TL can be used to manipulate FE and create stable, non-trivial FE topological polarization textures associated with the FE displacement field.

In-situ Bragg coherent diffractive imaging experiment

Figure 2. In-situ Bragg coherent diffractive imaging experiment. (a) Coherent x-rays (red) are incident on a nanoflake (yellow) containing polarization vortices and FE domains subjected to twisted UV light (Red-yellow-blue) illumination. A schematic of a Dion-Jacobson layered oxide CBNO crystal which has been shown to harbor FE with high Curie temperature and large in-plane polarization is used as a model system to study twisted light control of atomic displacements, microscopic FE polarization and domain structure with near-atomic resolution observed as (b) shifts and broadening of the recorded Bragg peak intensity. Simulated phase of twisted UV lights with topological charges (ℓ\ell) of 1 and −1-1, carrying OAM quantized in units of ℓℏ\ell\hbar per photon (c). In our experiments, the diffracted X-rays collected at the detector plane carry information about the 3D electron density and atomic displacement fields within the nanoflake that allows us to monitor recorded X-ray intensity variations near the Bragg spot during a cyclical application of light topological charge from ℓ=0\ell=0, 1, −1-1, and back to 0.

Results

To conduct the in-situ TL BCDI experiment, we first synthesized free-standing CBNO 2D nanoflakes using a polymer precursor method [36]. These nanoflakes were then transferred onto the sample holder (see Materials and Methods and Supplemental Data, Figs. S1 and S2).

TL, also known as optical vortices or light carrying OAM, consists of electromagnetic fields [28, 39] with one or more singularities where the phase [28] is undefined, causing the intensity to vanish. The discovery of TL beams in visible light produced in free-space [28, 40] has generated significant interest across various scientific disciplines, including quantum information processing [41], optical tweezers, super-resolution microscopy [42], and telecommunications [43].

Figure 2(c) illustrates exemplary phase and 2D intensity profiles (see Supplemental Data, Figs. S4 and S29) in planes perpendicular to the propagation direction (z-direction) for the Laguerre-Gaussian (LGℓ_\ell) mode of TL used in this experiment. LG beams are solutions to the paraxial wave equation [28], and in the lowest order of the paraxial approximation, the electric and magnetic fields, as well as the vector potential, are purely transverse. Using the Lorenz gauge, the longitudinal component of the electric field (see Supplemental Data, S7 and Figs. S4, S5, S6, and S7) can be expressed in terms of a scalar potential as:

Ez(r,t)=−∂tΦ(r,t)=−cei(kz−ωt)A0⋅∇⊥Π(r),(1)E_z(\mathbf{r},t)=-\partial_t\Phi(\mathbf{r},t)=-c e^{i(kz-\omega t)}\mathbf{A}_0\cdot\nabla_{\perp}\Pi(\mathbf{r}), \tag*{(1)}

where EzE_z is the longitudinal component of the electric field, Φ\Phi is the scalar potential, cc is the speed of light, kk is the wave number, ω\omega is the angular frequency, A0\mathbf{A}_0 is the vector potential amplitude, ∇⊥\nabla_{\perp} is the transverse gradient, and Π(r)\Pi(\mathbf{r}) is the complex amplitude of the LG beam, expressed as Π(r)=Π0(r,z)exp⁡(iℓϕ)\Pi(\mathbf{r})=\Pi_0(r,z)\exp(i\ell\phi).

In our in-situ BCDI setup, we utilized a spatial light modulator (SLM) (more details in Materials and Methods) to focus a 375 nm continuum (with constant beam power over time) of LG beams with varied topological charge ℓ\ell normal to the in-plane direction ([100]-direction) of our CBNO sample (see Figure 1). We generated a spatially inhomogeneous TL electric field EOAM\mathbf{E}_{\mathrm{OAM}} and measured its intensity profile. For full derivation, see Supplementary Data, Figs. S4, S5, and S6 and Note S7.

EOAM(ρ,t)={E(2ρw)∣ℓ∣e−ρ2w2(cos⁡(ℓϕ−ωt)ex−σsin⁡(ℓϕ−ωt)ey),for 0≤t≤T,0,for t>T.(2)\mathbf{E}_{\mathrm{OAM}}(\rho,t)= \begin{cases} E\left(\frac{\sqrt{2}\rho}{w}\right)^{|\ell|}e^{-\frac{\rho^2}{w^2}}\left(\cos(\ell\phi-\omega t)\mathbf{e}_x-\sigma\sin(\ell\phi-\omega t)\mathbf{e}_y\right), & \text{for } 0\le t\le T,\\ 0, & \text{for } t>T. \end{cases} \tag*{(2)}

In equation (2), TT is the time during which the CBNO nanoflake is under LGℓ_\ell beam illumination. At the time t=0t=0, we switched on the TL UV beam and, with the aid of the SLM and the neutral density filter, selected TL electric field polarization (σ=0\sigma=0) components ex\mathbf{e}_x and ey\mathbf{e}_y and magnitude such that E∥[100]\mathbf{E}_{\parallel[100]} as shown in Figure 1. We aligned the X-ray detector and CBNO crystal to satisfy the Bragg condition for the d002d_{002} inter-atomic spacing of the (002) Bragg spot, as shown in Figure 3. For a duration TT, we measured 3D diffraction near the (002) peak for beam topological charge ℓ=0\ell= 0. This procedure was repeated cyclically for ℓ=0\ell= 0, 1, −1-1, and back to 0 (see Figure 3(a)). During the experiment, we observed TL-induced structural changes by tracking the characteristic asymmetrical behavior in diffraction patterns (see Figure 1(b) and Figure 3(a)).

Real-space topologically-protected polar structures under a vortex-light field

Figure 3. Real-space topologically-protected polar structures under a vortex-light field. (a) Two-dimensional slices from the recorded three-dimensional coherent diffraction patterns for the (002) Bragg peak, used to measure the time evolution of the (b) FE displacement field during twisted light-induced structural changes. The polarization vortex core is identified at time t0t_0 under LGℓ_\ell beam illumination of ℓ=0\ell= 0 (zero OAM). As time progresses to t3t_3, changes in the displacement field are observed due to the nucleation of a vortex-antivortex pair. (c) A two-dimensional map showing the evolution of the central cut of a Bloch point in the freestanding CBNO. (d) The magnitude of the toroidal moment density under LGℓ_\ell beam of ℓ=0\ell= 0. (e) A rendering at time t0t_0 of CBNO under a TL electric field of ℓ=0\ell= 0, illustrating a vortex spanning the volume of the crystal along the zz−direction. Each cross-section perpendicular to the zz− direction contains a vortex texture, and the cut through the yy− direction shows a bloch-like behavior around the vortex core. The scale bar in (a) represents 0.01 Å−1^{-1} while that in (b), (c), and (d) represents 60 nm.

X-rays scattered by a single CBNO nanoflake satisfying the (002) Bragg condition were recorded on an area detector (see Figure 3(a) and Supplemental Data, S11). The experimental geometry, combined with the random orientation of the exfoliated nanoflakes, ensured that the (002) Bragg reflections from separate particles were well separated, allowing individual reflections to be isolated on the detector. The nanoflake was kept under continuous coherent TL UV illumination before the imaging experiment. From the coherent diffraction data, we reconstructed both the 3D distribution of the displacement field and the electron density (ρ(x,y,z)\rho(x,y,z)), along [2], u002(x,y,z)u_{002}(x,y,z), in an individual nanoflake with 14 nm spatial resolution (see Supplemental Data Figs. S14, S15, and S16).

Figure 3(b) shows a cross-section of the 3D displacement field [u002(x,y,z=z0)u_{002}(x,y,z = z_0)] in the FE nanoflake. The [2] direction is approximately along the zz−axis, while the X-ray beam is almost parallel to the xx−axis. Depending on the topological polar structure type, the angular distribution will have distinct features associated with a topological invariant known as the topological FE charge winding number:

Q=18π∫dA εijk∑καZκ,α∗uκα(r)∣∑καZκ,α∗uκα(r)∣⋅[∂j(∑κβZκ,β∗uκβ(r)∣∑κβZκ,β∗uκβ(r)∣)×∂k(∑κγZκ,γ∗uκγ(r)∣∑κγZκ,γ∗uκγ(r)∣)].(3)Q = \frac{1}{8\pi} \int dA \, \varepsilon_{ijk} \frac{\sum_{\kappa\alpha} Z^{*}_{\kappa,\alpha}u_{\kappa\alpha}(r)}{\left|\sum_{\kappa\alpha} Z^{*}_{\kappa,\alpha}u_{\kappa\alpha}(r)\right|} \cdot\left[ \partial_j \left( \frac{\sum_{\kappa\beta} Z^{*}_{\kappa,\beta}u_{\kappa\beta}(r)}{\left|\sum_{\kappa\beta} Z^{*}_{\kappa,\beta}u_{\kappa\beta}(r)\right|} \right) \times\partial_k \left( \frac{\sum_{\kappa\gamma} Z^{*}_{\kappa,\gamma}u_{\kappa\gamma}(r)}{\left|\sum_{\kappa\gamma} Z^{*}_{\kappa,\gamma}u_{\kappa\gamma}(r)\right|} \right) \right]. \tag*{(3)}

This is a quantized value associated with FE defects such as vortices or skyrmions that measure the number of times the displacement vector field (or related order parameter) wraps around a given point or region in space. These topological features can influence the electronic properties and are important for applications in topological quantum computing and advanced electronic devices [13, 11]. In Eqn.(3) AiA_i is the surface of a small volume containing the polar structure, ϵijk\epsilon_{ijk} is the Levi-Civita symbol, and Zκ,iα∗Z^{*}_{\kappa,i\alpha} is the Born effective charge tensor component for atom κ\kappa, relating the ii-th component of the polarization to the α\alpha-th component of the displacement of atom κ\kappa, obtained from DFT computations (see Materials and Methods and

Supplemental Data S9, and Figs. S8, S9, S10, and S11 for the measured displacement field in Figure 3(b). Displacement fields generated by FE vortices are characterized by a winding number Q=12π∮C∇θ⋅dl=+1Q=\frac{1}{2\pi}\oint_C\nabla\theta\cdot d\mathbf{l}=+1 around the vortex core in 2D (Supplementary Section S12), regardless of their polarity (clockwise or anticlockwise) [44]. The angle θ=arctan⁡(δu002δxu002)\theta=\arctan\left(\frac{\delta u_{002}}{\delta x u_{002}}\right) is related to the components of the displacement vector u002(r)\mathbf{u}_{002}(r). For polar BPs, the winding number captures the change in the displacement (polarization) direction in three dimensions. For anti-polar Bloch points (ABPs), the winding number captures the reversal in the polarization direction. The integral remains the same but yields a negative winding number due to the opposite orientation. Merons and anti-merons are characterized by half-integer winding numbers Q=±14π∮C∇θ⋅dlQ=\pm\frac{1}{4\pi}\oint_C\nabla\theta\cdot d\mathbf{l} and a discontinuous polarization field.

By inspecting Figure 3, we identified topological polar structures such as vortices, merons, and BPs in the displacement field and electron density maps obtained by BCDI. Figure 3(b) shows that at time t0t_0, two types of antiphase domains, α\alpha and β\beta, can form in CBNO due to the presence of two bifurcation options. Combined with two possible in-plane polarization directions [36, 29] ("+" being parallel to the aa and bb-axes and "−" being antiparallel to them), there are a total of four antiphase-FE Z2×Z4\mathbb{Z}_2 \times\mathbb{Z}_4 domains (α+\alpha+, β−\beta-, α−\alpha-, and β+\beta+) existing in this system under ℓ=0\ell=0 (see Supplemental Figure S17).

Unlike trimerization observed in other improper FE materials such as the hexamanganites and hexaferrites [45], where three-fold symmetric domains are formed, the transition in CBNO leads to an in-plane FE domain structure shown in Figure 3.

In Figure 3(b) and Figure 3(d), we qualitatively analyze the polarization vortex by inspection of the displacement field. Approximately 50 nm higher, in-plane Bi-ion displacement due to LGℓ_\ell electric field of ℓ=0\ell=0 induces a periodic strain modulation (Figure 3) in the xyxy-plane above the vortex and creates tensile strain due to its larger lattice constant. However, approximately 50 nm below, the material compensates for this by creating a predominantly compressive strain with a smaller lattice constant. By looking at multiple planes above and below along the cc-direction, within the spatial resolution of the experiment, we corroborate this periodic modulation of tensile and compressive strain in the xyxy-plane. The strain maps also show intersecting domains that form fork-like patterns in the yzyz-plane. These fork-like features not only guarantee the formation of the vortices but also account for the observed polar BPs (Figure 3(d,Figure 3e)).

The magnitude of the LGℓ_\ell electric field EOAM=10E_{\mathrm{OAM}}=10 kV/cm for ℓ=0\ell=0 is smaller than the reported coercive field [29] for CBNO (see Supplemental Data, Section S7, Table S1, Fig. S6), thus explaining why we do not observe significant FE domain switching. To investigate how the observed strain modulation affects the FE polarization, we measured the relative shifts of α\alpha and β\beta domains (Supplemental Data, Figs. S17 and S18) under the TL induced electric field. We notice that at time t1t_1 under an OAM field of ℓ=1\ell= 1, the α+\alpha^{+} and β+\beta^{+} domains shift by vectors of 1/5[110]1/5[110] and −1/5[110]-1/5[110], respectively, relative to the β−\beta^{-}. This corresponds to polar displacements as a result of Bi3+\mathrm{Bi}^{3+} shift in the range of 0.25–0.6 Å, which is consistent with what is found in bulk CBNO [36].

The evolution of the complex polar textures shown in the planar views of the CBNO displacement field and magnitude (Figure 3) under the vortex electric field EOAM\mathbf{E}_{\mathrm{OAM}} can be better assessed by quantitative inspection of the three-dimensional BCDI reconstructions. Besides the winding number QQ, we can describe the topological structure in terms of a non-zero toroidal moment [15, 16, 17], defined as G=12N∑i=1Nri×Pi\mathbf{G} = \frac{1}{2N}\sum_{i=1}^{N}\mathbf{r}_{i} \times\mathbf{P}_{i}, where Pi\mathbf{P}_{i} is the local dipole moment located at ri\mathbf{r}_{i} and NN is the number of dipoles (cells). Pi\mathbf{P}_{i} is estimated using the Born effective charge tensor, Pi=∑καZκ,iα∗uκα\mathbf{P}_{i} = \sum_{\kappa\alpha} Z^{*}_{\kappa,i\alpha}\mathbf{u}_{\kappa\alpha}.

We compute the FE toroidal moment from the reconstructed polarization. Regions of the large toroidal moment are plotted in Figure 4(a), where several ‘tubes’ and loops corresponding to the cores of vortices, antivortices, BPs, and ABPs are visible (see Supplemental Data, Figs. S23, S24, S25, S26, and S27). Unlike in incompressible fluids, where the divergence must vanish, a non-zero divergence of the FE polarization due to complex topological structures like FE vortices and BPs signifies the presence of bound charges that play a role in the formation and stability of these structures under the TL-induced vortex field. BPs and ABPs are identified by positive (red) and negative (blue) values within the reconstructed toroidal moment field. We observe a large number of three-dimensional loops (Figure 4(a)) that resemble vortex rings under LGℓ_{\ell} electric field of ℓ=1\ell= 1. We consider the case of one such loop, identified by plotting an isosurface corresponding to a maximum threshold of ±0.05Ps\pm0.05P_s (Figure 4(a)). This loop is located in the vicinity of a single Bloch (anti-Bloch) point, spanning a region of the CBNO crystal. This loop texture represents a departure from the coaxial string-like topology observed under the TL-induced vortex field of ℓ=0\ell= 0 and ℓ=1\ell= 1 (Supplemental Data, Figs. S25, S26, and S27). The 2D cut planes in Figure 4(b) show the vector field of change in PmagP_{\mathrm{mag}} (see Supplemental Data Section S16), allowing us to identify topological features such as head-to-head convergence, tail-to-tail divergence, BPs, ABPs, and merons.

Twisted light-induced polar Bloch point transitions and strain evolution

Figure 4. Twisted light-induced polar Bloch point transitions and strain evolution. (a) BCDI rendering of the magnitude of the toroidal moment density under OAM of ℓ=−1\ell=-1, capturing the topology formed by Bloch and anti-Bloch points. (b) 2D slices depicting the density of BPs and ABPs in the toroidal field, marked by planes in (a), with the normalized magnitude of polarization ΔPmag\Delta P_{\mathrm{mag}} vectors overlaid. (c) Hysteresis-like loop capturing OAM field-induced strain εzz\varepsilon_{zz} and BP transition states in CBNO. (d) BCDI strain map under an OAM field of ℓ=0\ell=0 showing that BPs are associated with regions of reduced polarization magnitude or significant strain gradients. Scale bars correspond to 100 nm.

In Ref. [26], polar BPs were observed in strained PbTiO3_3 films using phase-field simulations and aberration-corrected scanning transmission electron microscopy at the locations where a vortex core intersected a domain wall. Similarly, we find that the BP pair is located at the intersection of the vortex-antivortex loop with a domain wall separating regions of opposite strain magnitude (Figure 4(d)). It was predicted in the same reference that polar BPs can be stabilized using substrate-induced tensile strain in FE films. In Ref. [44], a periodic shear strain landscape was generated by stacking freestanding FE perovskite layers with controlled twist angles, akin to Moiré lattices [46]. An important remark is that in both cases above, the inter-layer strain in the film cannot be dynamically tuned.

Here, in Figure 4(c), we observe that under LGℓ_{\ell} electric field transitioning from ℓ=0\ell= 0 to ℓ=1\ell= 1, then ℓ=−1\ell= -1, and back to ℓ=0\ell= 0, we can manipulate the global and local strain within the CBNO crystal in a controllable manner (Supplemental Data, Figs. S19 and S20). Under a global tensile strain ϵzz\epsilon_{zz} of 0.7%, there is a transition from BPs to merons. A slight increase in the tensile strain to ϵzz\epsilon_{zz} of 0.75%, by changing the vortex field to ℓ=−1\ell= -1, reverts to a global Bloch point state. Subsequent compressive strain back to ℓ=0\ell= 0 transitions to a global ABP state within the CBNO crystal (see Supplemental Data, Tables S2 and S3 and Fig. S22). Figure S20 shows the local mapping of the strain modulations and induced shear strains that mediate the observed BP transitions. It is worth noting that such controlled strain manipulation and the observed strain landscape are unique, as they cannot be attained, to the best of our knowledge, either by epitaxial strain or by any pattern of externally applied stresses or electrodes.

In Figs. 3 and 4, we have presented experimental evidence of the formation of polar topological states in the quasi-2D FE nanoflake, with emerging polarization vortex structures that are axially symmetric at time t0t_0 under LGℓ_{\ell} electric field of ℓ=0\ell= 0. However, the creation of a vortex costs additional energy due to the existence of the vortex core where the FE is suppressed. Thus, under UV LGℓ_{\ell} beam illumination of ℓ=1\ell= 1, the singularity at the vortex core can be completely eliminated by the escape of the vector field into the third dimension (see Fig. S19) along the vortex axis, leading to the emergence of chiral meron-like structures in the flake. Under TL illumination of ℓ=−1\ell= -1, the system returns to a BP state, while subsequent compressive strain allows the polar structure to transition to an ABP state under a TL-induced LGℓ_{\ell} field of ℓ=1\ell= 1.

In addition to the observed TL-induced OAM field-induced strain and manipulation of real-space FE (FE) polar structures from BCDI, OAM-dependent Raman measurements on CBNO suggest a possible correlation with twisted UV light-induced excitation of the zone-center and zone-boundary modes in CBNO (see details in Materials and Methods and Supplemental Data, Figs. S9, S10 and S12).

Figure 5(a) illustrates the experimental setup for OAM-dependent Raman measurements on CBNO. Details of the experimental methodology can be found in Materials and Methods. A reference scan was initially performed to confirm the unperturbed crystal structure of CBNO, as shown in Supplemental Data Fig. S29. Subsequent Raman scans were conducted under various OAM conditions, similar to BCDI measurements (ℓ=0\ell= 0, ℓ=1\ell= 1, ℓ=−1\ell= -1, and returning to ℓ=0\ell= 0). We monitored both the in-plane zone-center phonon mode, reported around 190 cm−1^{-1} for the Bi-O bond [30], and the out-of-plane phonon mode, as depicted in Figure 5(d) and Figure 5(c), respectively. Illumination with ℓ=0\ell= 0 resulted in tensile strain along the c-axis, while the OAM field with ℓ=1\ell= 1 induced a compressive effect. These findings are consistent with the BCDI measurements. A reversible behavior was observed with ℓ=−1\ell= -1, and returning to ℓ=0\ell= 0 resulted in a Raman spectral shift that matched the initial state. This reversibility was also seen in the zone-boundary phonon mode, which has been reported around 580 cm−1580\ \mathrm{cm}^{-1} for the Nb-O in-plane Raman peak [35], as shown in Figure 5(b). The consistent nature of these trends was confirmed by tracking the Raman peak position at full width at half maximum (FWHM), as illustrated in Figure 5(e).

Twisted-light dependent Raman spectroscopy on CsBiNb$_2$O$_7$

Figure 5. Twisted-light dependent Raman spectroscopy on CsBiNb2O7CsBiNb_2O_7: (a) Schematic of the in-situ Raman experimental setup designed to probe photo-induced structural changes, modulation of phonon modes, and domain dynamics in a CBNO nanoflake under LGℓ_\ell light illumination. The setup utilizes a 532 nm Raman laser with a 375 nm coherent twisted light of varying topological charges. (b) Line plots showing the behavior of the out-of-plane phonon mode for the Bi-O layer. The spectra demonstrate reversible shifts in peak positions, characterized by a tensile shift from the reference to ℓ=0\ell= 0, compression under ℓ=1\ell= 1, and a reversion to a tensile state under ℓ=−1\ell= -1, with minimal changes upon returning to ℓ=0\ell= 0. This suggests significant OAM-induced modulation of vibrational states. (c) Line plots illustrating the out-of-plane phonon modes of the Bi-O layer, including isolated Bi atom responses. The plots indicate compression effects attributed to thermal influences from continuous illumination, highlighting the sensitivity of Bi phonon modes to UV LGℓ_\ell light and thermal input. (d) Line plots for the in-plane phonon modes of the Bi-O layers, showing reversible Raman peak shifts similar to those observed in the out-of-plane Bi-O modes, underscoring the symmetric impact of OAM on in-plane Bi-O vibrations. (e) Summary of Raman peak shifts across different modes when interacting with twisted light carrying various topological charges. Insets depict the structural and bond modifications along with torsional strain effects induced by twisted light, demonstrating the impact of OAM on lattice dynamics (refer to Supplemental Data, Figs. S10, S12, S31, and S32 for detailed analysis). (f–g) Schematic representations of interactions between twisted light with topological charges ℓ=1\ell= 1 (Compressive Strain) and ℓ=−1\ell= -1 (Tensile Strain), with Bi-O tetrahedra and Nb-O octahedra, respectively. These illustrate the torsional and structural changes imparted by non-zero OAM, contributing to the understanding of vortex-light-induced Raman activity.

Schematics showing the effects of twisted light (TL)-induced OAM fields (ℓ=1\ell= 1 and ℓ=−1\ell= -1) on the Bi-O and Nb-O bonds are presented in Figure 5(f) and Figure 5(g), respectively. The schematic is inspired by the data shown in Figure 5(b),(Figure 5c) & (Figure 5d), where we observe a reversible tensile and compressive behavior in strain in the phonon modes. This reiterates and emphasizes the ability of the TL-induced electric fields to control the bending and breathing of the phonon modes in the nanocrystal. An inset illustrating the inverse case is shown in Figure 5(e), and a detailed analysis of the shifts in polarization is depicted in Supplemental Data, Fig. S9. The Raman spectroscopy results align with the proposed interaction mechanism between TL and FE polarization in CBNO. Qualitatively, comparisons between BCDI and Raman spectroscopy reveal a hysteresis-like behavior in both strain and Raman peak shifts. This is analogous to the traditional hysteresis observed between polarization and electric field in FEs, where the beam topological charge acts similarly to a variable electric field. However, the complex nature of the LGℓ\ell electric field provide a wider space to manipulate the sample. This demonstrates the potential for controlling FE topologies using TL, opening new avenues for research into vortex-light–matter interactions.

We propose that the interaction mechanism involves the coupling of TL with the electronic structure, leading to shifts in phonon modes within the CBNO nanoflake. One potential interaction mechanism is intra-band transitions, as discussed in Supplemental Data, S8 and S10. We suggest that electron-mediated shifts in the phonon mode structure result in the observed changes in strain, where a preferential shift in the electronic structure is noted (see Supplemental Data, Fig. S13) within the tight-binding approximation of the CBNO band structure. In both BCDI and Raman spectroscopy experimental techniques, the effects are observed post electron-phonon renormalization, leading to the observed Raman peak shifts, FE polarization, and strain behavior.

Our experimental results, supported by theoretical analysis, reveal that the OAM states ℓ=0\ell= 0 and ℓ=0′\ell= 0' correspond to a uniform electric field distribution, while ℓ=1\ell= 1 and ℓ=−1\ell= -1 result in a non-uniform electric field when illuminating the nanoflake. In both cases, the strength and direction of the electric field vary depending on the beam waist and the topological charge. Under a homogeneous electric field corresponding to ℓ=0\ell= 0, a stable two-dimensional vortex structure was observed at time t=0t = 0, with the vortex core aligned along one of the crystallographic axes [1] (see Figure 3(c)). A small quasi-static field applied at t=0t = 0 stabilized an axial vortex core, which is topologically protected by two antiparallel spontaneous nanodomains with nearly homogeneous spontaneous polarization (see Supplemental Data Section S8 and Fig. S21).

Similar to ferromagnetic vortices, the core of the observed ferroelectric vortex behaves like a dipole, offering the potential to rotate the vortex axis under the influence of a small quasi-static electric field, whether homogeneous or inhomogeneous. Due to strong ferroelectric anisotropy, which hinders the rotation of the dipolar vortex core and tends to align it along one of the crystallographic directions, the orientation of the core in the field is determined by the minimization of dipolar and anisotropy energies. As the vortex core is coupled to the antiphase domains, the plane of the vortex, which is perpendicular to the core axis, can rotate along with the core. The core size is sensitive to variations in the TL-induced electric field, and as these variations occur, the nanoflake gradually transitions toward a single-domain state, as observed at time t=t1t = t_1.

Manipulating FE polarization in quasi-2D materials like CsBiNb2_2O7_7 using incident vortex fields generated by coherent twisted ultraviolet light offers distinct advantages over using pulsed terahertz infrared photons, as traditionally employed in 3D perovskite materials like SrTiO3_3. Twisted ultraviolet light carries topological charge, imparting angular momentum to the electrons and ions in the material, leading to precise interactions with ferroelectric polarization. This helical phase can stabilize transitions between Bloch points and merons by aligning polarization vectors and modifying the energy landscape, thereby lowering the energy barriers for these transitions. The higher spatial precision and photon energy of UV light allow for more effective and localized polarization changes compared to terahertz infrared pulses. Additionally, vortex fields carrying orbital angular momentum can selectively interact with specific phonon modes or electronic states that are not easily accessible with terahertz infrared pulses, providing a unique tool for manipulating ferroelectric polarization.

In 3D perovskite oxides systems [5, 4, 7], pulsed terahertz infrared photons can resonantly excite specific phonon modes, inducing transient changes in ferroelectric polarization and numerous emergent phenomena [2, 3, 47]. However, the additional degree of freedom and increased pathways for energy dissipation in 3D systems make it more challenging to achieve the same level of control and precision as in quasi-2D systems. Therefore, using electric field of coherent twisted ultraviolet light to manipulate ferroelectric polarization in quasi-2D CsBiNb2_2O7_7 provides superior control and effectiveness, enabling the stabilization of complex topological transitions.

Conclusion and Outlook

We have demonstrated that the orbital angular momentum electric field from twisted light can dynamically induce strains (both tensile and compressive) up to ∼0.8%\sim0.8\% along the cc-axis in individual quasi-2D ferroelectric CsBiNb2O7CsBiNb_2O_7 crystals. These strains are quantifiable, with nanometer resolution, and can be controlled reversibly, creating significant strain gradients. This phenomenon is potentially universal to quasi-2D ferroelectric systems with weak out-of-plane bonding along the stacking direction. For example, in CsBiNb2O7CsBiNb_2O_7, the reported piezoelectric coefficient d33d_{33} (8 pC N−1^{-1}) [31] is significantly smaller than in typical 3D perovskite oxides (e.g., PbTiO3PbTiO_3, BaTiO3BaTiO_3 [21], BiFeO3BiFeO_3). Our observations show that topologically protected polar structures and transitions from Bloch points to merons occur under a light-carrying orbital angular momentum electric field, mediated by induced tensile strains. These findings align with earlier reports of polar Bloch points in strained epitaxial thin films. However, in epitaxial thin films, the mechanical boundary conditions are rigid and determined by the atom-on-atom replication of the substrate structure, leaving little room for modification. Additionally, reversible and dynamic control of torsional, shear, or inhomogeneous strain patterns is not easily achievable via epitaxy.

In summary, we have found that it is possible to induce non-trivial ferroelectric textures and polar structures, as well as strains in quasi-2D freestanding ferroelectric crystalline flakes. The driving force is the coupling between the twisted light’s spatially varying electric field and the crystal, which effectively imposes electromechanical (stress and vectorial electric field) boundary conditions. These couplings induce large strains and strain gradients in the ferroelectric material, resulting in vortex-like and Bloch-type modulations of the homogeneous polarization state. Furthermore, the nature of these 2D and 3D topological structures can be largely tuned by controlling the twisted light’s topological charge.

Moreover, the interaction of twisted light with ferroelectrics potentially opens new avenues for research into the dynamical twist-multiferroics [47] and the braiding of polar 1D structures. Upgrades in the coherence [48, 49] of synchrotrons and the development of time-resolved orbital angular momentum pump orbital angular momentum-probe Bragg coherent diffractive imaging techniques that require twisted X-rays [50] could potentially provide deeper insights into the dynamical control of novel states of matter that are unattainable by conventional thermodynamic methods. Additionally, the development of new Bragg coherent diffractive imaging methods to access in-plane strain and orbital angular momentum-dependent Raman spectroscopy to capture chiral or more nuanced twisted-light interactions could represent a paradigm shift in engineering beyond 2D. This includes the potential for dynamical manipulation of the twist angle in which other modes of twisted light, such as higher-order Bessel beams with higher penetration, could play important roles. Shedding further light on these advances would be a great challenge, for both theory and experiment, but is beyond the scope of the present work.

Materials and Methods

Fabrication of freestanding quazi-2D flakes

Free-standing CBNO 2D flakes were synthesized using a molten-salt assisted method [36, 35]. Further, pre-characterization was carried out to confirm the stoichiometry using XPS (Supplemental Data, S4), morphology using XRD and SEM (Supplemental Data, S2 and S3) and bandgap using transmission spectroscopy (Supplemental Data, S5).

Twisted light Bragg coherent diffractive imaging experiments

The Si (111) monochromator at the Advanced Photon Source’s sector 34-ID-C was utilized to isolate coherent X-ray photons with an energy of 9.0 keV. The beam’s monochromaticity was defined by an energy bandwidth of 1 eV, and it exhibited a 0.7 μm transverse coherence length. The X-ray beam was focused onto the sample using a pair of Kirkpatrick-Baez mirrors situated after the beam-defining aperture. For this experiment, the beam size was set to 700 nm by 700 nm. The principle of the Bragg coherent diffraction experiment is illustrated in Fig. S3. A Medipix2 CMOS X-ray detector was positioned around the diffraction sphere using a motorized arm. The detector was aligned with the outgoing (002) characteristic Bragg reflection from the CBNO sample. To magnify the interference fringes in the diffraction pattern, the detector was placed 1.2 m away from the sample. An evacuated flight tube positioned in the path from the sample to the detector helped minimize photon scattering losses in the air. For such a photon-starving technique as nanoscale Bragg coherent diffractive imaging, the use of an evacuated flight tube, high sensor gain, and the detector’s photon counting mode are essential. An Ultra-Violet laser of 375nm, guided via a confocal system available at the 34-IDC beamline, was focused down to 10 μm at a normal incidence to the isolated CBNO flake of interest. During the collection of rocking curves, the sample was continuously illuminated. We employed a spatial light modulator (SLM) to select the optical beam’s topological charge, thereby controlling the OAM and torque transferred to the crystal (see Supplemental Data, S6 and Fig. S3). Rocking curves were collected as a series of 2D diffraction patterns near the CBNO 002 Bragg peak, corresponding to 2θ=13.26∘2\theta= 13.26^\circ, with a scanning range of Δθ=±0.2∘\Delta\theta= \pm0.2^\circ in the vicinity of the Bragg peak. A total of 360 patterns were gathered in a single rocking curve. The dataset for the virgin state (topological charge, ℓ=0\ell= 0) was obtained before cycling the CBNO crystal. The subsequent states under continuous LGℓ_\ell beam illumination, ℓ=1\ell= 1, ℓ=−1\ell= -1 and ℓ=0\ell= 0 were recorded after 30 cycles of ℓ=0\ell= 0 illumination and release. This takes about 30 mins for one single 3D scan to be collected. Thus for all OAM applications T=t0=t1=t2=t3=30T = t_0 = t_1 = t_2 = t_3 = 30 mins. This ensured that the system was entirely in equilibrium under laser illumination.

Phase retrieval process

The collected 3D diffraction patterns were inverted from reciprocal into real space using iterative phase retrieval algorithms [21]. This gives us the complex wavefield post diffraction, which is given by ρ(r)=ρ0(r)eiϕ(r)\rho(\mathbf{r}) = \rho_0(\mathbf{r})e^{i\phi(\mathbf{r})} where ϕ(r)\phi(\mathbf{r}) is the phase and ρ(r)\rho(\mathbf{r}) is the amplitude of the wave. The phase carries the information of about the displacement field u002u_{002}, which can be determined by using ϕ(r)=G002.u002(r)\phi(\mathbf{r}) = \mathbf{G}_{002}.\mathbf{u}_{002}(\mathbf{r}). The shape and size of the nanocrystal are estimated from the reconstructed Bragg Electronic Density(ρ(r)\rho(\mathbf{r})). For a given applied TL topological charge, real-space images of the CBNO nanocrystal were reconstructed with approximately 33 nm spatial resolution as determined by the phase retrieval transfer function (PRTF) (see Supplemental Data, Fig. S2B). The 3D nature of the reconstructed nanocrystal allows us to slice through the particle and analyze the different FE topologies. Fienup’s hybrid input-output (HIO) approach was the basis for iterative phase retrieval techniques, which were further enhanced by randomized overrelaxation. When the measurement locations are close enough to one another to satisfy the oversampling criterion, an important stage in the process is to reverse the diffraction data using a computer algorithm. The first stage is to assume a three-dimensional support volume where the sample density will be completely limited. These techniques impose a Fourier transform, both forward and backward, between the real and reciprocal spaces, imposing an intensity mask constraint in the former and a support constraint in the latter.

Twisted-light Raman spectroscopy

A 532 nm wavelength light from the Witec Alpha 300R series was used to perform the Raman Measurements. The Witec Alpha 300R was modified with a free beam coupler to introduce the twisted light of wavelength 375 nm on to the sample surface. During the collection of a single spectrum, the sample was continuously illuminated. We employed a spatial light modulator (SLM) to select the optical beam’s topological charge, thereby controlling the OAM and torque transferred to the crystal (see Supplemental Data, S18 and Fig. S29). An Andor CCD was used to collect and count photons scattered from the CBNO nanoflake. A spectroscope grating of 1800 gr/mm was used to obtain a high-resolution spectrum of the sample under various topological charges. A Zeiss 100x microscope objective was used for focusing the laser beams onto the nanoflake as shown in Fig. S29. The integration time of 1 s was used to collect 100 accumulations of the spectrum to obtain a clean spectrum with a low signal-to-noise ratio. The reference spectrum was collected before cycling the system. The subsequent spectrums were collected under continuous illuminations of ℓ=1\ell= 1, ℓ=−1\ell= -1 and ℓ=0\ell= 0 were collected.

First-principles calculations

We performed first-principles calculations using the CASTEP (Cambridge Serial Total Energy Package) [51] and Quantum ESPRESSO (QE) [52] packages. The calculations involving the application of an external light beam carrying Orbital Angular Momentum (OAM) with different topological charges (ℓ\ell) were performed using CASTEP. The experimental electric field component of the OAM beam was explicitly included in these calculations. We employed the Perdew-Burke-Ernzerhof (PBEsol) exchange-correlation functional within the generalized gradient approximation (GGA) [53] for all calculations. We used norm-conserving pseudopotentials and a plane-wave basis set with a kinetic energy cutoff of 600 eV. The pseudopotentials for Cs, Bi, Nb, and O were taken from the CASTEP library.

The Brillouin zone was sampled using a 6×6×66 \times6 \times6 Monkhorst-Pack grid [54]. The self-consistent field (SCF) calculations were converged to an energy threshold of 1×10−61 \times10^{-6} eV. The initial molecular structure of CsBiNb2_2O7_7 was obtained from experimental theoretical data predictions available on the Materials Project. The structures were optimized using density functional theory (DFT) as implemented in Quantum ESPRESSO. The Perdew-Burke-Ernzerhof (PBEsol) exchange-correlation function was chosen to describe the electron-electron interactions. The unit cell parameters are: a=11.87380a = 11.87380 Å, b=5.45730b = 5.45730 Å, c=5.55750c = 5.55750 Å, with angles α=90∘\alpha= 90^\circ, β=90∘\beta= 90^\circ, γ=90∘\gamma= 90^\circ. Atomic positions were fully relaxed until the forces on each atom were less than 0.01 eV/Å, and the stress tensor was below 1×10−31 \times10^{-3} GPa. The polarizability was computed using density functional perturbation theory (DFPT) [55]. Infrared and Raman spectra were obtained by calculating the dynamical charges and the Raman tensors, respectively. The dielectric function and the Raman activities were computed to simulate the IR and Raman spectra. For Reviews only, highlighted references (optional) Please select 5—10 key references and provide a single sentence for each, highlighting the significance of the work.

Acknowledgements

E.F. and M.N conceived the project and conceptualized the experiments. J.S. and J.J. grew and provided free-standing and epitaxial 2D CBNO samples. N.N, V.T, J.B, W.C,R.H and E.F Performed the OAM BCDI measurements and characterization. BCDI analysis using phase retrieval algorithms was performed by N.N and E.F OAM Raman measurements and analysis was performed by N.N, V.T, P.B, A.N, M.N, and E.F., A.D performed the DFT computations. N.N and E.F. wrote the paper. All the authors discussed the results and commented on the manuscript. We thank S. Williams (RPI), A. Ndao (UCSD), L. Caretta (Brown University), and D. Kumah (Duke University) for the stimulating conversations on this manuscript.

Funding: This work was supported by the US Department of Energy (DOE), Basic Energy Sciences, Materials Sciences and Engineering Division, under grant No. DE-SC0023148 (OBCDI technique, ORAMAN and analysis), and the US Department of Defense, Air Force Office of Scientific Research (AFOSR), under award No. FA9550-23-1-0325 (Program Manager: Dr. Ali Sayir) for work on probing topological vortices and piezoelectric enhancements. J.S. thanks the support from NSF under award number 2024972. This research used resources of the Advanced Photon Source (APS), a U.S. Department of Energy (DOE Office of Science User Facility, operated for the DOE Office of Science by Argonne National Laboratory (ANL) under contract No. DE-AC02-06CH11357. Author contributions: R.H., W.C., and E.F. designed and performed the BCDI experiment. J.J. and J.S. grew and provided the samples. V.T, P.B., and M.N synthesised the twisted beam.M.N and V.T acknowledge the support of The Office of Advanced Scientific Computing Research within the Office of Science US Department of Energy (DOE) under grant No DE-SC0024676. All authors interpreted the results and contributed to writing the manuscript. Competing interests: The authors declare no competing interests. Data and materials availability: The data reported in this paper are archived at Beamline 34-ID-C at the Advanced Photon Source at Argonne National Lab and are available upon request. All code, including the reconstruction algorithm, is also available upon request.

Data availability

The experimental and DFT data are available upon reasonable request.

Competing interests

The authors declare no competing interests.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information (optional)

References

  1. [1]von Hoegen, A., Mankowsky, R., Fechner, M., Först, M. & Cavalleri, A. Probing the interatomic potential of solids with strong-field nonlinear phononics. Nature 555, 79–82 (2018).DOI
  2. [2]Liu, M. et al. Terahertz-field-induced insulator-to-metal transition in vanadium dioxide metamaterial. Nature 487, 345–348 (2012).DOI
  3. [3]Tobey, R., D下一ist, A., Pyon, S., Takayama, T. & Takagi, H. Light-induced superconductivity in a stripe-ordered cuprate. Science 331, 189–191 (2011).DOI
  4. [4]Li, X. et al. Terahertz field–induced ferroelectricity in quantum paraelectric srtio3. Science 364, 1079–1082 (2019).DOI
  5. [5]Pancaldi, M. et al. Terahertz electric-field driven dynamical multiferroicity in srtio _3. arXiv preprint arXiv:2210.01690 (2022).DOI
  6. [6]Gao, L., Prokhorenko, S., Nahas, Y. & Bellaiche, L. Dynamical control of topology in polar skyrmions via twisted light. Phys. Rev. Lett. 132, 026902 (2024).DOI
  7. [7]Gao, L., Prokhorenko, S., Nahas, Y. & Bellaiche, L. Effective gyration of polar vortex arrays controlled by high orbital angular momentum of light. Phys. Rev. B 109, L121110 (2024).DOI
  8. [8]Kittel, C. Physical theory of ferromagnetic domains. Rev. modern Phys. 21, 541 (1949).DOI
  9. [9]Kittel, C. Theory of the structure of ferromagnetic domains in films and small particles. Phys. Rev. 70, 965 (1946).DOI
  10. [10]Tang, Y. et al. Observation of a periodic array of flux-closure quadrants in strained ferroelectric pbtio3 films. Science 348, 547–551 (2015).DOI
  11. [11]Ma, J. et al. Controllable conductive readout in self-assembled, topologically confined ferroelectric domain walls. Nat. nanotechnology 13, 947–952 (2018).DOI
  12. [12]Karpov, D. et al. Three-dimensional imaging of vortex structure in a ferroelectric nanoparticle driven by an electric field. Nat. Commun. 8 (2017).DOI
  13. [13]Yadav, A. et al. Observation of polar vortices in oxide superlattices. Nature 530, 198–201 (2016).DOI
  14. [14]Das, S. et al. Observation of room-temperature polar skyrmions. Nature 568, 368–372 (2019).DOI
  15. [15]Wang, Y. et al. Polar meron lattice in strained oxide ferroelectrics. Nat. Mater. 19, 881–886 (2020).DOI
  16. [16]Zhang, Q. et al. Nanoscale bubble domains and topological transitions in ultrathin ferroelectric films. Adv. Mater. 29, 1702375 (2017).DOI
  17. [17]Luk’Yanchuk, I., Tikhonov, Y., Razumnaya, A. & Vinokur, V. Hopfions emerge in ferroelectrics. Nat. Commun. 11, 2433 (2020).DOI
  18. [18]Salje, E. & Scott, J. Ferroelectric bloch-line switching: A paradigm for memory devices? Appl. Phys. Lett. 105 (2014).DOI
  19. [19]Morozovska, A. N. et al. Electric field control of three-dimensional vortex states in core-shell ferroelectric nanoparticles. Acta Materialia 200, 256–273 (2020).DOI
  20. [20]Karpov, D. et al. Nanoscale topological defects and improper ferroelectric domains in multiferroic barium hexaferrite nanocrystals. Phys. Rev. B 100, 054432 (2019).DOI
  21. [21]Shi, X. et al. Enhanced piezoelectric response at nanoscale vortex structures in ferroelectrics. arXiv preprint arXiv:2305.13096 (2023).DOI
  22. [22]Behera, P. et al. Electric field control of chirality. Sci. advances 8, eabj8030 (2022).arxiv.org/abs/2105.14109
  23. [23]Shafer, P. et al. Emergent chirality in the electric polarization texture of titanate superlattices. Proc. Natl. Acad. Sci. 115, 915–920 (2018).DOI
  24. [24]Lovesey, S. W. & van der Laan, G. Resonant x-ray diffraction from chiral electric-polarization structures. Phys. Rev. B 98, 155410 (2018).DOI
  25. [25]Chauleau, J.-Y. et al. Electric and antiferromagnetic chiral textures at multiferroic domain walls. Nat. materials 19, 386–390 (2020).DOI
  26. [26]Wang, Y.-J. et al. Polar bloch points in strained ferroelectric films. Nat. Commun. 15, 3949 (2024).DOI
  27. [27]Jeong, L. J. e. a. Y. Y., Chaehwa. Revealing the three-dimensional arrangement of polar topology in nanoparticles. Nat. Commun. 10, 1038/s41467-024-48082-x (2024).
  28. [28]Allen, L., Beijersbergen, M. W., Spreeuw, R. & Woerdman, J. Orbital angular momentum of light and the transformation of laguerre-gaussian laser modes. Phys. review A 45, 8185 (1992).
  29. [29]Fang, X., Hu, R., Wang, Y., Huang, F.-T. & Cheong, S.-W. Hybrid improper ferroelectricity in highly cleavable single crystals of dion-jacobson-compound csbinb2o7. Adv. Electron. Mater. 8, 2100929 (2022).DOI
  30. [30]Fennie, C. J. & Rabe, K. M. Ferroelectricity in the dion-jacobson csbinb2o7 from first principles. Appl. physics letters 88 (2006).DOI
  31. [31]Chen, C. et al. Ferroelectricity in dion-jacobson abinb 2 o 7 (a= rb, cs) compounds. J. Mater. Chem. C 3, 19–22 (2015).DOI
  32. [32]Goff, R. J. et al. Leakage and proton conductivity in the predicted ferroelectric csbinb2o7. Chem. Mater. 21, 1296–1302 (2009).DOI
  33. [33]Snedden, A., Knight, K. S. & Lightfoot, P. Structural distortions in the layered perovskites csanb2o7 (a= nd, bi). J. Solid State Chem. 173, 309–313 (2003).DOI
  34. [34]Hervoches, C., Irvine, J. T. S. & Lightfoot, P. Two high-temperature paraelectric phases in sr 0.85 bi 2.1 ta 2 o 9. Phys. Rev. B 64, 100102 (2001).
  35. [35]Jiang, J. et al. Giant pyroelectricity in nanomembranes. Nature 607, 480–485 (2022).DOI
  36. [36]Guo, Y. et al. Unit-cell-thick domain in free-standing quasi-two-dimensional ferroelectric material. Phys. Rev. Mater. 5, 044403 (2021).DOI
  37. [37]Robinson, I. & Harder, R. Coherent x-ray diffraction imaging of strain at the nanoscale. Nat. materials 8, 291 (2009).DOI
  38. [38]Goff, R. J. et al. Leakage and proton conductivity in the predicted ferroelectric csbinb2o7. Chem. Mater. 21, 1296–1302, DOI: 10.1021/cm8030895 (2009). https://doi.org/10.1021/cm8030895.
  39. [39]Forbes, A. Controlling light’s helicity at the source: orbital angular momentum states from lasers. Philos. Transactions Royal Soc. A: Math. Phys. Eng. Sci. 375, 20150436 (2017).DOI
  40. [40]Wang, G. et al. Free-space creation of a perfect vortex beam with fractional topological charge. Opt. Express 31, 5757–5766 (2023).DOI
  41. [41]Erhard, M., Fickler, R., Krenn, M. & Zeilinger, A. Twisted photons: new quantum perspectives in high dimensions. Light. Sci. & Appl. 7, 17146–17146 (2018).DOI
  42. [42]Shen, Y. et al. Optical vortices 30 years on: Oam manipulation from topological charge to multiple singularities. Light. Sci. & Appl. 8, 90 (2019).DOI
  43. [43]Tran, V. et al. On the exploration of structured light transmission through a multimode fiber in a reference-less system. AP Photonics 8 (2023).DOI
  44. [44]Sánchez-Santolino, G. et al. A 2d ferroelectric vortex pattern in twisted batio3 freestanding layers. Nature 626, 529–534 (2024).
  45. [45]Das, H., Wysocki, A. L., Geng, Y., Wu, W. & Fennie, C. J. Bulk magnetoelectricity in the hexagonal manganites and ferrites. Nat. communications 5, 2998 (2014).DOI
  46. [46]Lau, C. N., Bockrath, M. W., Mak, K. F. & Zhang, F. Reproducibility in the fabrication and physics of moiré materials. Nature 602, 41–50 (2022).DOI
  47. [47]Basini, M. et al. Terahertz electric-field-driven dynamical multiferroicity in srtio3. Nature 1–6 (2024).
  48. [48]Kerby, J. The advanced photon source upgrade: A brighter future for x-ray science (2023).DOI
  49. [49]Fohtung, E., Karpov, D. & Baumbach, T. Bragg coherent diffraction imaging techniques at 3rd and 4th generation light sources. Mater. Discov. Des. By Means Data Sci. Optim. Learn. 203–215 (2018).DOI
  50. [50]Nazirkar, N. P., Shi, X., Shi, J., N’Gom, M. & Fohtung, E. Coherent diffractive imaging with twisted x-rays: Principles, applications, and outlook. Appl. Phys. Rev. 11 (2024).DOI
  51. [51]Segall, M. et al. First-principles simulation: ideas, illustrations and the castep code. J. physics: condensed matter 14, 2717 (2002).DOI
  52. [52]Giannozzi, P. et al. Quantum espresso: a modular and open-source software project for quantum simulations of materials. J. physics: Condens. matter 21, 395502 (2009).DOI
  53. [53]Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. review letters 77, 3865 (1996).pubmed.ncbi.nlm.nih.gov/10062328
  54. [54]Monkhorst, H. J. & Pack, J. D. Special points for brillouin-zone integrations. Phys. review B 13, 5188 (1976).DOI
  55. [55]Baroni, S., De Gironcoli, S., Dal Corso, A. & Giannozzi, P. Phonons and related crystal properties from density-functional perturbation theory. Rev. modern Phys. 73, 515 (2001).DOI

Paper details

vlnc:W_7ptzty4n
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 8

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents