Optical fiber end face integrated nonlinear activated metasurface and processing method and application thereof

Abstract

The application provides a kind of optical fiber end face integrated nonlinear activation super surface and its processing method and application, comprising the following steps: the optical fiber end face is pretreated to form adhesion promotion layer on the optical fiber end face;Unsolidified photoresist is coated on the adhesion promotion layer of the optical fiber end face;The three-dimensional model of nonlinear activation super surface is constructed;According to the three-dimensional model of nonlinear activation super surface, the optical fiber end face photoresist is exposed to two-photon polymerization to form a nonlinear activation super surface with multiple sub-super surfaces on the optical fiber end face;Gradient development and drying are carried out to remove unexposed photoresist;The application effectively solves the problem that the optical fiber end face is difficult to integrate complex wiener structure, and the prepared nonlinear activation super surface structure is stable, combined closely with the optical fiber end face, and is convenient for popularization and application.

Description

Technical Field

[0001] This invention belongs to the field of optical fiber end-face processing technology, specifically relating to a nonlinear activated metasurface integrated into an optical fiber end-face, its processing method, and its application. Background Technology

[0002] In the context of fiber endface functionalization, achieving nonlinear activation at the micro- and nano-scale has become crucial for enhancing the intelligence and response performance of optoelectronic devices. However, due to space constraints and the limited nonlinear response strength of materials, traditional methods struggle to achieve controllable and significant nonlinear effects at the endface. To address this, nonlinear activated metasurfaces have emerged. These are multilayered three-dimensional structures constructed from arrays of nano- or subwavelength artificial structures. Their core feature lies in the adaptive adjustment of optical responses (such as phase delay, transmittance, or reflectivity) with incident light intensity, enabling intelligent control of the light field distribution. This demonstrates significant application potential in cutting-edge fields such as biomimetic visual perception, weak light enhancement detection, and high-power laser limiting protection.

[0003] Traditional fiber end-face processing techniques have significant limitations in fabricating such complex structures: although electron beam lithography has nanometer precision, its processing efficiency is extremely low, making it difficult to fabricate multilayer three-dimensional structures with high aspect ratios; focused ion beam milling has a slow fabrication speed and is prone to causing lattice damage that affects light transmission; and traditional laser direct writing technology is limited by the diffraction limit and cannot achieve subwavelength level fine structure design.

[0004] Two-photon polymerization (TPP) technology, with its breakthrough diffraction limit resolution, true 3D processing capabilities, and material compatibility, has opened up new avenues for the fabrication of fiber end-face metasurfaces. However, achieving high-performance, large-size, and multi-level coupled nonlinear activated metasurfaces still faces the following severe challenges: First, the design and modeling of massive pixelated units. Nonlinear activated metasurfaces consist of tens of thousands of subwavelength pixel pillars, and the height of each unit needs to be precisely modulated according to a complex phase matrix. Traditional mechanical modeling software cannot handle such large-scale phase data mapping, making it difficult to automate the design and fabrication process. Second, controlling splicing errors under large field-of-view processing. The diameter of the full cladding of optical fibers is typically 125 μm or larger, exceeding the single-scan field of view of high-magnification, high-numerical-aperture two-photon direct-write systems. When processing the entire end face, a splicing strategy must be adopted, but the "jump error" at the splicing point of the multilayer structure can easily lead to phase discontinuities, severely degrading the optical performance of the metasurface. Third, the stability and stress accumulation of high aspect ratio multilayer structures. To achieve nonlinear activation, multilayer stacked structures with a total height of hundreds of micrometers are typically required. At such high aspect ratios, the shrinkage stress of the polymer during curing can cause the structure to warp, collapse, or even peel off from the smooth end face of the optical fiber; at the same time, the submicron-level vertical alignment accuracy between layers is also difficult to guarantee.

[0005] Therefore, how to stably fabricate high-precision, multi-layered metasurfaces with nonlinear adaptive functions at the fiber end face is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] In view of the shortcomings of the prior art, the purpose of this invention is to provide a nonlinear activated metasurface integrated on the fiber end face, its processing method and application, so as to achieve high-precision integration of a nonlinear activated metasurface with nonlinear adaptive function on the fiber end face.

[0007] To achieve the above and other related objectives, this invention provides a method for fabricating a nonlinear activated metasurface integrated into an optical fiber endface. The nonlinear activated metasurface comprises multiple spaced sub-metasurfaces, each sub-metasurface including a substrate and an array of nanopillars disposed on the substrate; the substrates of adjacent sub-metasurfaces are connected by connecting pillars. The method includes the following steps:

[0008] The fiber end face is pretreated to form an adhesion-promoting layer on the fiber end face;

[0009] Uncured photoresist is coated onto the adhesion-promoting layer on the fiber end face;

[0010] Construct a three-dimensional model of a nonlinear activated metasurface;

[0011] Based on the three-dimensional model of the nonlinear activated metasurface, the photoresist on the fiber end face is subjected to two-photon polymerization exposure processing to form a nonlinear activated metasurface with multiple sub-metasurfaces on the fiber end face.

[0012] Gradient development and drying are performed to remove unexposed photoresist.

[0013] The method for constructing the three-dimensional model of the nonlinear activated metasurface includes:

[0014] An inverse neural network is used to solve the target output light field in reverse order to obtain the target phase matrix of each sub-metasurface in the nonlinear activated metasurface; the target phase matrix is ​​the phase delay φ(x,y) required by the nanopillars at different positions (x,y) on the substrate of the sub-metasurface.

[0015] Based on the linear mapping relationship between column height and phase delay, the column height H(x,y) of each nanopillar in each sub-metasurface is calculated.

[0016] Based on the coordinates and height of each nanopillar in each sub-metasurface, the preset substrate size, and the preset coordinates and size of the connecting pillars, a three-dimensional model of the nonlinear activated metasurface is generated using script-based automated modeling; the three-dimensional model is in STL format.

[0017] Preferably, the method for obtaining the target phase matrix includes:

[0018] A forward physical propagation model is constructed using scalar diffraction theory or full-wave electromagnetic simulation algorithm; a randomly generated initial phase matrix is ​​input into the forward physical propagation model to calculate the corresponding light intensity distribution image; the light intensity distribution image obtained from the initial phase matrix is ​​converted into a wavefront amplitude intensity distribution to establish a phase matrix-wavefront amplitude intensity distribution dataset.

[0019] An inverse neural network is constructed, and the wavefront amplitude intensity distribution is used as input and the phase matrix is ​​used as output. The constructed inverse neural network is trained by minimizing the loss function between the predicted phase matrix output by the inverse neural network and the actual phase matrix input to the inverse neural network. The weights of the inverse neural network are iteratively updated.

[0020] The grayscale image of the target output light field is converted into the target wavefront amplitude intensity distribution, and the target wavefront amplitude intensity distribution is input into the trained inverse neural network to obtain the target phase matrix.

[0021] Preferably, the pretreatment of the optical fiber end face includes surface activation treatment and silane coupling agent grafting; the surface activation treatment includes oxygen plasma treatment and / or ultraviolet ozone treatment to introduce hydrophilic functional groups such as hydroxyl and carboxyl groups into the optical fiber end face to improve surface energy.

[0022] Preferably, the silane coupling agent used for grafting is 3-(trimethoxysilyl)propyl methacrylate. In this way, the methoxy (-OCH3) group of the silane coupling agent undergoes hydrolytic condensation with the hydroxyl group on the fiber end face to form a strong Si-O-Si covalent bond; simultaneously, the terminal methacrylate group extends outward, providing active sites for subsequent photoresist polymerization.

[0023] Preferably, the pretreatment of the fiber end face further includes ultrasonic cleaning; the ultrasonic cleaning process involves ultrasonically cleaning the fiber end face in acetone, ethanol and ultrapure water for 5 minutes each, and then drying it with nitrogen; in this way, organic pollutants, inorganic dust particles and other impurities adsorbed on the fiber end face can be removed, improving cleanliness and avoiding affecting the adhesion between the subsequent nonlinear activated metasurface and the fiber end face.

[0024] Preferably, the gradient development includes:

[0025] The fiber end face with nonlinear activated metasurface was immersed in propylene glycol methyl ether acetate for 18 min.

[0026] The fiber end face with the nonlinear activated metasurface was transferred and immersed in isopropanol for 2 to 3 minutes.

[0027] Preferably, after gradient development and drying, the end face of the optical fiber with nonlinear activated metasurface is cured at low temperature. This allows the unreacted active groups in the nonlinear activated metasurface to be further crosslinked, thereby improving its degree of polymerization, mechanical strength and long-term chemical stability.

[0028] The present invention also provides a nonlinear activated metasurface integrated on the fiber end face, wherein the nonlinear activated metasurface is obtained by the above-described processing method.

[0029] Preferably, the height of the nonlinear activated metasurface is 120 μm to 350 μm, and the interlayer spacing is 50 μm to 100 μm; the maximum lateral envelope diameter of the nanopillar array in the nonlinear activated metasurface is 15 μm to 250 μm.

[0030] The present invention also provides an application of the above-mentioned nonlinear activated metasurface in environmental monitoring or biological detection.

[0031] As described above, the nonlinear activated metasurface integrated into the fiber end face, its processing method, and its application provided by this invention have the following beneficial effects:

[0032] This invention activates the fiber end face through oxygen plasma treatment and / or ultraviolet ozone treatment, and constructs molecular bridges using silane coupling agent grafting technology. This allows the methoxy group (-OCH3) at one end of the silane coupling agent to undergo a hydrolytic condensation reaction with the hydroxyl group (-OH) on the fiber end face, forming a stable Si-O-Si covalent bond. Meanwhile, the organic functional group (such as methacrylate) at the other end of the silane coupling agent forms a molecular-level adhesion-promoting layer that can cross-link with the photoresist molecular chains. This significantly increases the interfacial bonding force between the fiber end face and the photoresist, effectively reducing the risk of the nonlinear activated metasurface detaching from the fiber end face. Simultaneously, the two-photon polymerization process is used to expose the photoresist coated on the fiber end face, enabling high-precision fabrication of the nonlinear activated metasurface. This allows for highly efficient, nonlinear optical field manipulation in the reflection or transmission spectrum of the optical fiber.

[0033] This invention employs a gradient development method to remove unexposed photoresist, addressing substrate swelling and structural collapse / damage issues of the nonlinear activated metasurface during development, thus ensuring the bonding stability between the nonlinear activated metasurface and the fiber end face. Furthermore, the low-temperature curing following development allows for further cross-linking of unreacted active groups in the nonlinear activated metasurface, thereby improving its degree of polymerization, mechanical strength, and long-term chemical stability, ultimately enhancing the reliability of the nonlinear activated metasurface in use.

[0034] The processing method of this invention has the advantages of low processing control difficulty, high processing accuracy, and high yield. The nonlinear activated metasurface processed by it has the advantages of structural integrity and tight bonding with the fiber end face, which facilitates its promotion and application. Attached Figure Description

[0035] Figure 1 This is a flowchart illustrating a method for fabricating a nonlinear activated metasurface integrated on the fiber end face according to an embodiment of the present invention.

[0036] Figure 2 This is a schematic diagram of a nonlinear activated metasurface integrated on the end face of an optical fiber, provided in an embodiment of the present invention.

[0037] Figure 3 A scanning electron microscope image of a nonlinear activated metasurface integrated at the end face of an optical fiber, provided in an embodiment of the present invention.

[0038] Figure 4 An exploded view of a nonlinear activated metasurface integrated into an optical fiber endface according to an embodiment of the present invention.

[0039] Figure 5 for Figure 3 The image shows an electron microscope (EM) image of the surface morphology of the neutron metasurface of the nonlinear activated metasurface.

[0040] Figure 6 This is a comparison of the output light intensity changes between bare optical fiber and optical fiber with integrated nonlinear activated metasurface at input optical power ranging from 0.02mW to 1mW.

[0041] Explanation of reference numerals in the attached figures

[0042] Fiber 1, nonlinear activated metasurface 2, sub-metasurface 2a, substrate 21, nanopillar array 22, connecting pillar 23. Detailed Implementation

[0043] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification.

[0044] Please see Figures 1 to 6 It should be understood that the structures, proportions, sizes, etc., illustrated in the accompanying drawings are merely for illustrative purposes to aid those skilled in the art and to facilitate understanding and reading. They are not intended to limit the scope of the invention and therefore have no substantial technical significance. Any modifications to the structure, changes in proportions, or adjustments to size, without affecting the effectiveness and purpose of the invention, should still fall within the scope of the technical content disclosed in this invention. Furthermore, the terms such as "upper," "lower," "left," "right," "middle," and "one" used in this specification are merely for clarity and not intended to limit the scope of the invention. Changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of the invention's implementation.

[0045] like Figure 1 As shown, the fabrication method for the nonlinear activated metasurface integrated into the fiber end face provided by the present invention includes:

[0046] S1. Pre-treat the fiber end face to form an adhesion-promoting layer on the fiber end face.

[0047] It should be noted that the optical fiber can be single-mode or multi-mode, and there is no limitation on this. In this embodiment, the optical fiber is preferably a single-mode optical fiber of type SMF-28, with a core diameter of 105 μm, a cladding diameter of 125 μm, and an operating wavelength λ of 532nm~1550nm.

[0048] Once the optical fiber is selected, if the smoothness and cleanliness of the fiber end face are poor, the fiber end face needs to be polished and cleaned with high precision to ensure the smoothness and cleanliness of the fiber end face.

[0049] In one optional embodiment, the cleaning process includes ultrasonic cleaning. The specific steps of ultrasonic cleaning are as follows: the fiber end face is ultrasonically cleaned in acetone, ethanol and ultrapure water for 5 to 10 minutes each, and then dried with nitrogen gas to remove organic pollutants, inorganic dust particles and other impurities adsorbed on the fiber end face, so as to avoid affecting the adhesion between the nonlinear activated metasurface made by photoresist and the fiber end face.

[0050] In a preferred embodiment, the pretreatment of the optical fiber end face includes surface activation treatment and silane coupling agent grafting; wherein the surface activation treatment includes oxygen plasma treatment and / or ultraviolet ozone treatment.

[0051] The specific steps of oxygen plasma treatment are as follows: the cleaned fiber end face is placed in an oxygen plasma cleaner for treatment. The treatment parameters are: RF power 50 W, treatment time 40 seconds. During this process, the fiber end face is bombarded with high-energy oxygen plasma, which can achieve both physical etching, creating nanoscale pits on the originally smooth inert surface to increase the specific surface area, and chemical modification, generating active hydrophilic groups such as hydroxyl (-OH) and carboxyl (-COOH) groups on the fiber end face. Contact angle measurements showed that the water contact angle of the fiber end face was reduced to below 35°, indicating that the surface energy of the fiber end face treated with oxygen plasma was significantly improved, laying the foundation for subsequent chemical bonding.

[0052] The specific steps of ultraviolet ozone treatment are as follows: the cleaned optical fiber end face is placed in an ultraviolet ozone cleaner for treatment. The treatment parameters are: ultraviolet lamp power 15W, treatment time 20 minutes, to generate active hydrophilic groups such as hydroxyl (-OH) on the optical fiber end face. Tests using a contact angle meter showed that the water contact angle of the optical fiber end face was reduced to below 35°, indicating that the surface energy of the optical fiber end face treated with ultraviolet ozone was significantly improved, laying the foundation for subsequent chemical bonding.

[0053] Silane coupling agent grafting refers to the process by which a hydrolyzable alkoxy group (such as trimethoxysilyl) at one end of a silane coupling agent undergoes a hydrolytic condensation reaction with a hydroxyl group at the end face of an optical fiber, forming a stable covalent bond (such as Si-O-Si), while the organic functional group (such as methacrylate) at the other end of the silane coupling agent extends outward to form an adhesion-promoting layer that can cross-link with the photoresist molecular chain. To ensure grafting uniformity, this embodiment preferably uses a gas-phase method for silane coupling agent grafting.

[0054] The silane coupling agent can be any silane coupling agent suitable for photoresist systems, such as methacryloxysilane or vinylsilane. There is no limitation on this, as long as the selected silane coupling agent is compatible with the selected photoresist. In this embodiment, the preferred silane coupling agent is 3-(trimethoxysilyl)propyl methacrylate, which is methacryloxysilane with the model number KH570.

[0055] For ease of understanding, the specific steps of vapor phase grafting are described below: The surface-activated fiber end face is rapidly (preferably within 5 minutes) transferred to a sealed glass desiccator, and a petri dish containing an ethanol solution of 3-(trimethoxysilyl)propyl methacrylate at a volume ratio of 0.8%–1.2% (preferably 1%) is placed at the bottom of the desiccator. The mixture is then allowed to stand at room temperature for 30 minutes to complete the vapor phase grafting. During this process, the methoxy (-OCH3) group at one end of the 3-(trimethoxysilyl)propyl methacrylate undergoes hydrolytic condensation with the hydroxyl groups on the fiber end face, forming a strong Si-O-Si covalent bond. Simultaneously, the methacrylate group at the other end of the 3-(trimethoxysilyl)propyl methacrylate extends outward, forming a dense molecular-level adhesion-promoting layer on the fiber end face, providing stable interfacial bonding for the subsequent multilayer structure formed by photoresist polymerization.

[0056] This invention significantly increases the interfacial bonding force between the fiber end face and the subsequently coated photoresist by performing surface activation treatment and silane coupling agent grafting on the fiber end face. This reduces the possibility of the multilayer structure prepared by the photoresist detaching from the fiber end face during subsequent development or use, and provides the possibility for stable integration of optical fiber with high-precision multilayer structure.

[0057] S2. Coat the adhesion-promoting layer on the end face of the optical fiber with uncured photoresist.

[0058] It should be noted that the photoresist can be any existing negative photoresist for two-photon printing, such as IP-DIP2, YF-TPP, MON-DIP, or ATE-DIP (produced by Yantai Magic Nanotechnology), and there is no limitation on the type, as long as it is compatible with the selected silane coupling agent; in this embodiment, IP-DIP2 photoresist is preferred.

[0059] In step S2, uncured photoresist is spin-coated onto the end face of the optical fiber using a spin coater to form a uniform, defect-free photoresist layer on the end face of the optical fiber. The thickness of the photoresist layer is determined according to the actual situation; in this embodiment, the thickness of the photoresist layer is preferably 3 μm.

[0060] After spin coating of photoresist, the fiber end face with photoresist can be placed in a 45°C oven for 12 minutes to remove solvent residue. This not only improves the adhesion of photoresist to the fiber end face, but also releases the internal stress generated by spin coating, making the formed photoresist layer smoother and more uniform.

[0061] S3. Construct a three-dimensional model of the nonlinear activated metasurface.

[0062] like Figure 2 and Figure 4 As shown, the nonlinear activation metasurface 2 is a multilayer stacked structure, which includes multiple sub-metasurfaces 2a spaced apart in the height direction (i.e., the stacking direction), and adjacent sub-metasurfaces 2a are connected by connecting pillars 23. The sub-metasurface 2a includes a substrate 21 and a nanopillar array 22 disposed on top of the substrate 21. The substrate 21 is used to provide a flat processing plane. The nanopillar array 22 is composed of a large number of closely arranged nanopillars, and the height of each nanopillar is not the same, so as to facilitate the realization of phase compensation from 0 to 2π by using the metasurface propagation phase principle. In this way, after the incident light field passes through the phase compensation of each nanopillar array and the diffraction accumulation of multiple sub-metasurfaces, it can activate nonlinear control effects such as adaptive amplitude limiting or wavefront conversion in the target space.

[0063] It should be noted that the number of neutron metasurfaces 2a in the nonlinear activated metasurface 2 is determined according to the actual situation and is not limited thereto, as long as it can adaptively respond to changes in incident light and dynamically control the target output light field; in this embodiment, the overall height of the nonlinear activated metasurface is 120 μm - 350 μm to ensure the structural stability of the prepared nonlinear activated metasurface.

[0064] It should be noted that the thickness of the substrate 21 needs to meet the mechanical strength requirements, and is generally not less than 0.5 μm; in this embodiment, the thickness of the substrate 21 is 0.5 μm.

[0065] It should be noted that the horizontal cross-section of the substrate 21 is square, and its cross-sectional dimensions are determined based on the dimensions of the end-face fiber diffraction network determined by the target output light field. As long as the dimensions of the substrate 21 are greater than the corresponding dimensions of the end-face fiber diffraction network, the end-face fiber diffraction network is represented by a nanopillar array 22.

[0066] The size of the end-face fiber diffraction network is determined by the inverse neural network algorithm. Specifically, it is determined by the total number of neurons designed in the inverse neural network algorithm and the size of a single neuron (nanopillar). Regarding the number of neurons, in order to avoid overfitting due to an excessive number of neurons, the array range of neurons (i.e., nanopillars) is generally 50×50~200×200.

[0067] It should be noted that the cross-sectional shape of the neuron (i.e., nanopillar) is square, and its side length is generally λ / 2~3λ. Too small or too large a dimension will lead to an increase in the system's diffraction efficiency and size. Since the working wavelength λ is 532 nm~1550 nm, the calculated side length of a single neuron (nanopillar) is 266 nm~4650 nm. For ease of subsequent modeling, this is rounded down, resulting in a side length of 300 nm~4 μm for a single neuron (nanopillar). Furthermore, considering the field of view of the two-photon printing process, the cross-sectional size of the end-face fiber diffraction network (i.e., nanopillar array 22) is 15 μm×15 μm~250 μm×250 μm (i.e., the maximum lateral envelope diameter of the nanopillar array 22 is 15 μm~250 μm). In this case, the substrate size only needs to be larger than the determined nanopillar array size.

[0068] At this point, the array range of the nanopillars in the nanopillar array 22, the side length of the cross-section, and the structural dimensions of the substrate 21 on which the nanopillars are located can be determined.

[0069] Furthermore, the height of each nanopillar in the nanopillar array 22 is determined by the target phase matrix determined by the target output light field.

[0070] It should be noted that the position and number of connecting posts 23 between two adjacent sub-metasurfaces 2a can be determined according to the actual situation, as long as the multiple connecting posts 23 are uniformly distributed around the vertical centerline of the sub-metasurface 2a. The cross-section of the connecting post 23 is preferably circular, and the height of the connecting post 23 (i.e., the interlayer spacing of the nonlinear activation metasurface) needs to be determined by comprehensively considering three dimensions: optical design, functional requirements, and process feasibility. Specifically, based on the Rayleigh-Sommerfeld diffraction theory, the evolution of the light field under different propagation distances is scanned through numerical simulation to select the spacing range that can achieve ideal diffraction modulation and ensure effective transmission of interlayer signals. For the working wavelength λ of 532nm~1550nm and the determined cross-sectional size of a single neuron, the theoretical design value of the height of the connecting post 23 should not be less than 20 μm. Furthermore, considering the integration requirements of the device, a smaller interlayer spacing (i.e., the height of the connecting post) is better. However, since the interlayer spacing (i.e., the height of the connecting post) directly affects the intensity of the equivalent nonlinear activation, a balance needs to be struck between weak light enhancement and strong light limiting. This application uses the input-output S-shaped response curve as an evaluation index to screen out the interlayer spacing with the optimal nonlinear modulation effect, finding that the interlayer spacing should not exceed 300 μm. Finally, from the perspective of two-photon processing technology, it was found that if the interlayer spacing is too small, it will be difficult to remove the interlayer residue during development, resulting in easy adhesion; while if the interlayer spacing is too large, it will not only lead to excessive processing time, but also reduce the uniformity of upper layer curing. Therefore, actual verification shows that an interlayer spacing of 50 μm to 100 μm can ensure sufficient development, structural independence, and good processability. Therefore, the interlayer spacing (i.e., the height of the connecting post) is 50 μm to 100 μm; the cross-sectional diameter of the connecting post 23 only needs to meet the mechanical strength requirements, and is generally 2 μm to 20 μm.

[0071] At this point, the structural dimensions of connecting column 23 are determined.

[0072] In this step, the method for constructing the three-dimensional model of the nonlinear activated metasurface includes:

[0073] S31. The target output light field is solved in reverse by using an inverse neural network to obtain the target phase matrix of each sub-metasurface in the nonlinear activated metasurface; the target phase matrix is ​​the phase delay φ(x,y) required by the nanopillars at different positions (x,y) on the substrate of the sub-metasurface 2a.

[0074] In this process, the array range of nanopillars in each sub-metasurface 2a and the side length of the cross-section of the nanopillars can be determined, thereby determining the structural dimensions of the substrate 21.

[0075] S32. Based on the linear mapping relationship between column height and phase delay, calculate the column height H(x,y) of each nanopillar in each sub-metasurface;

[0076] According to the metasurface phase propagation mechanism, the phase delay φ of the nanopillar relative to the surrounding medium and the height H of the nanopillar satisfy the formula (1).

[0077]

[0078] Where H is the height of the nanopillar;

[0079] λ is the operating wavelength;

[0080] n re The refractive index of the nanopillar;

[0081] n en The refractive index of the environmental medium;

[0082] As can be seen from formula (1), the phase delay φ of the nanopillar relative to the environmental medium is linearly related to the height H of the nanopillar. Since the phase delay distribution φ(x,y) required by the nanopillar at different positions (x,y) on the substrate of the sub-supersurface 2a is known, the height H(x,y) of each nanopillar in the sub-supersurface 2a can be directly calculated.

[0083] Specifically, the height H(x,y) of each nanopillar (i.e., nanopillars at different positions) in the sub-supersurface is calculated by the following formula (2), which is derived from formula (1).

[0084]

[0085] Wherein, φ(x,y) is the phase delay required for the nanopillars at different positions (x,y) on the substrate by the sub-supersurface.

[0086] λ is the operating wavelength. In this embodiment, λ is 532nm ~ 1550nm.

[0087] n re The refractive index of the nanopillar is 1.618. In this embodiment, the nanopillar is made of IP-Dip2 photoresist.

[0088] n en The refractive index is 1 for the ambient medium. In this embodiment, the ambient medium is air, which has a refractive index of 1.

[0089] Since φ(x,y) varies between 0 and 2π, substituting φ(x,y) = 2π into formula (2) will allow us to calculate the maximum height H of the nanopillar. max = .

[0090] At this point, the position and structural parameters of each nanopillar in each sub-metasurface, as well as the structural parameters of substrate 21, are known.

[0091] It should be noted that the structural parameters of the substrate 21 in each sub-metasurface must be consistent.

[0092] The position and number of connecting posts 23 between two adjacent sub-supersurfaces 2a can be set according to the actual situation, as long as the multiple connecting posts 23 are evenly distributed with the vertical center line of the sub-supersurface 2a as the center; in this embodiment, the height of the connecting post 23 is 50μm ~ 100μm; the cross-section of the connecting post 23 is circular with a diameter of 2μm ~ 20μm.

[0093] Then, based on the coordinates and heights of each nanopillar in each sub-metasurface, the preset substrate size, and the preset coordinates and dimensions of the connecting pillars, a three-dimensional model of the nonlinear activated metasurface in STL format is directly generated using scripting languages ​​such as Python. The specific generation principle of the three-dimensional model of a single sub-metasurface is now explained using this example:

[0094] The script calculates the three-dimensional coordinates of each cube (i.e., the substrate 21 and each nanopillar) at its eight vertices based on the dimensions of the substrate 21, the coordinates of each nanopillar, the cross-sectional dimensions, and the height of the sub-supersurface. Then, it defines that each of the six faces of each cube is composed of two triangular facets, and traverses the six faces of each cube in a counterclockwise direction to obtain the vertex index of each triangular facet. Finally, it uses a mesh processing library to globally flatten and correct the normal vectors of the vertex and facet data of multiple discrete cubes, so as to seamlessly stitch the substrate 21 and multiple nanopillars into a single-layer topological model.

[0095] Based on the above method, a single-layer topological model of each sub-metasurface 2a can be constructed. Then, based on the size and coordinates of the connecting pillars 23 in two adjacent sub-metasurfaces 2a, a single-layer topological model of each connecting pillar layer is established. Finally, the topological models of each layer are topologically stitched together, and the stitched topological model is output as a high-resolution binary STL file to obtain a three-dimensional model of the nonlinear activated metasurface.

[0096] Furthermore, when the nonlinear activated metasurface is a large-sized structure with a diameter of not less than 100 μm, a script is needed to automatically divide the nonlinear activated metasurface into multiple sub-blocks adapted to the scanning field of view of the two-photon polymerization direct writing system, and calculate the coordinates of the 5 μm − 10 μm overlap area between adjacent sub-blocks to perform coordinate jump compensation at the splicing boundary, thereby eliminating the phase discontinuity problem caused by galvanometer jump error during subsequent two-photon polymerization processing. This invention completes the compensation of coordinate jump error at the splicing boundary during the model creation stage, so that the jump error at the splicing point can be offset during subsequent two-photon large field-of-view processing, ensuring the geometric continuity of the phase distribution between sub-blocks and improving the processing quality of the nonlinear activated metasurface.

[0097] Specifically, during the model creation phase, this invention employs a pre-compensation strategy to address the inherent sub-block splicing jump error in two-photon processing systems. First, the coordinate jump offset between adjacent sub-blocks is calibrated. Then, in the design model of each sub-block, a reverse offset correction is applied to the phase or structural coordinates near the splicing boundary, and the phase values ​​at the boundary are re-interpolated to ensure continuous phase distribution on both sides of the offset boundary. During processing, the actual positive jump error of the system cancels out the reverse pre-compensation in the model, thereby ensuring precise alignment of the physical splicing boundary, eliminating phase jumps, and ensuring the processing quality of the large field-of-view nonlinear activated metasurface.

[0098] S4. Based on the three-dimensional model of the nonlinear activated metasurface, the photoresist on the fiber end face is subjected to two-photon polymerization exposure processing to form a nonlinear activated metasurface on the fiber end face.

[0099] Specifically, step S4 includes:

[0100] S41. Import the three-dimensional model of the nonlinear activated metasurface into the two-photon polymerization direct writing system for layer slicing and processing to obtain the laser scanning path file.

[0101] It should be noted that the two-photon polymerization direct writing system is an existing technology. Specifically, one can choose the GT2 two-photon micro-nano 3D printing system launched by Nanoscribe, the MPO100 two-photon 3D printing equipment provided by Heidelberg GmbH, the D100 two-photon micro-nano processing system launched by Shenzhen Phoenix Technology Co., Ltd., or the Yuzhiquan two-photon three-dimensional direct writing lithography machine, etc. There is no limitation on this. In this embodiment, the GT2 two-photon polymerization direct writing system is preferred, and its built-in slicing software can perform layer slicing and processing on the imported three-dimensional model.

[0102] In this embodiment, the slicing direction is Z-axis.

[0103] The slice layer height is determined according to the actual situation; in this embodiment, the slice layer height is 0.15μm to ensure that there are enough processing layers in the Z-axis direction to form a smooth sidewall.

[0104] Different line fill types are set inside the slices obtained by slicing, such as spiral fill, straight fill, and cross fill. At the same time, the spacing between two adjacent scan lines in the same slice can be adjusted by setting the scan spacing.

[0105] The scanning spacing should be greater than the diameter of the laser focal spot and less than 1 / 50 of the diameter of the Airy disk. In this embodiment, the scanning spacing is preferably 0.12 μm to ensure appropriate overlap between adjacent scanning lines while taking efficiency into account, so as to facilitate the formation of a continuous and dense structure in subsequent processing.

[0106] S42. The two-photon polymerization direct writing system performs two-photon polymerization on the photoresist on the fiber end face according to the obtained laser scanning path file, so as to form a nonlinear activated metasurface on the fiber end face.

[0107] Zero-point positioning and printing parameter settings are performed before the formal scanning and exposure.

[0108] Zero-point positioning is an existing technology. Its specific steps are as follows: fix the optical fiber on the piezoelectric station of the two-photon polymerization direct writing system, observe the concave meniscus formed by the objective lens-gel interface using a 100x objective lens and a real-time CCD, and find the absolute zero point (Z=0) between the laser focal length and the substrate interface by interface leveling measurement and Z-axis fine adjustment.

[0109] Printing parameter settings: Set laser power P to 8 mW–15 mW and scanning speed v to 200000μm / s–300000 μm / s.

[0110] In two-photon polymerization, the transverse diameter D and longitudinal width L of a voxel are proportional to the laser power P and related to the scanning speed. The ratio is inversely proportional; to ensure exposure uniformity, the laser power and scanning speed linkage control function of the two-photon polymerization direct writing system is used during processing to dynamically adjust the laser power and scanning speed, keeping the exposure energy per unit volume stable. Simultaneously, the laser power and scanning speed are reduced 100µm from the starting point of the trajectory, 100µm before the ending point of the trajectory, and in the overlapping area. This effectively eliminates the "overexposure" phenomenon at the endpoints caused by the acceleration and deceleration of the pressure station during laser scanning, ensuring exposure uniformity, thus guaranteeing processing accuracy and preventing structural protrusions or stress concentrations at the splicing points.

[0111] S5. Perform gradient development and drying to remove unexposed photoresist.

[0112] The specific steps of gradient development in this process include:

[0113] S51. Immerse the end face of the optical fiber with the nonlinear activated metasurface horizontally into a glass dish containing propylene glycol methyl ether acetate (PGMEA) and let it stand for 18 minutes. During this time, you can gently shake the dish several times to promote the dissolution and diffusion of the uncrosslinked resin.

[0114] Propylene glycol methyl ether acetate, as a moderately polar solvent, can gently and effectively dissolve unreacted components of photoresist, while its swelling effect on the fiber end face is far less than that of strongly polar solvents such as acetone.

[0115] S52. The fiber end face with nonlinear activated metasurface is rapidly transferred to isopropanol (IPA) and immersed for 2 to 3 minutes; in this embodiment, immersion for 3 minutes is preferred.

[0116] Isopropanol is highly polar and can displace and remove residual propylene glycol methyl ether acetate and other trace residues. At the same time, due to its low surface tension, isopropanol effectively reduces the risk of damage to the nonlinear activated metasurface during the development process.

[0117] After gradient development, the fiber end face with nonlinear activated metasurface is dried. The preferred drying method is supercritical drying, which involves placing the fiber end face with nonlinear activated metasurface into the sample chamber of a supercritical carbon dioxide dryer, replacing the residual solvent with liquid CO2, and then slowly heating and pressurizing to above the critical point (e.g., 35°C, 80 bar), maintaining this state for 30 minutes, and finally slowly depressurizing to completely eliminate the gas-liquid interface and the resulting capillary forces, thus preventing the nonlinear activated metasurface from sticking, bending, or collapsing during the drying process.

[0118] In a further embodiment, the processing method for the nonlinear activated metasurface also includes: S6, low-temperature curing, which involves ultraviolet curing or thermal curing at a temperature below 60°C to obtain the final flexible optical waveguide.

[0119] To reduce the risk of thermal deformation of the flexible substrate, the preferred curing temperature is 40°C.

[0120] The specific steps of UV curing are as follows: the end face of the optical fiber with the nonlinear activated metasurface is placed under a 365 nm UV lamp with an intensity of 20 mW / cm² and irradiated in a 40°C drying oven for 20 minutes. This low-temperature post-curing process aims to further crosslink the unreacted active groups in the nonlinear activated metasurface, thereby improving its degree of polymerization, mechanical strength, and long-term chemical stability.

[0121] It should be noted that step S2 can be located between steps S41 and S42.

[0122] This invention also provides a nonlinear activated metasurface integrated with an optical fiber endface, which is obtained by the above-described method for fabricating a nonlinear activated metasurface integrated with an optical fiber endface. The obtained nonlinear activated metasurface is as follows: Figure 3 As shown, the scanning electron microscope image of the nonlinear activated metasurface neutron metasurface 2a is as follows. Figure 5 As shown.

[0123] The following is a comparison of the output light intensity changes of bare fiber and fiber end-face device (i.e., fiber with integrated nonlinear activated metasurface) under input optical power ranging from 0.02mW to 1mW: Figure 6 As shown. By Figure 6It is known that the output light intensity of a bare optical fiber is linearly related to the input light power, while the output light intensity of the fiber end-face device is nonlinearly related to the input light power. When the optical power input is low, the output light intensity of the fiber end-face device is higher than that of the bare optical fiber, thus enhancing the perception of weak light signals. When the optical power input is high, the output light intensity of the fiber end-face device is lower than that of the bare optical fiber, thus achieving optical limiting protection.

[0124] The present invention also provides an application of a nonlinear activated metasurface integrated at the fiber end face in environmental monitoring or biological detection.

[0125] In summary, the nonlinear activated metasurface proposed in this invention, constructed based on the fiber endface, achieves integrated optical field modulation structure and optical transmission channel compared to traditional planar micro / nano fabricated structures, possessing both high-efficiency optical coupling and micro / nano manipulation capabilities. Furthermore, this application modifies the fiber endface and combines two-photon polymerization and gradient development techniques to significantly improve the interfacial bonding strength and structural stability between the fiber endface and the prepared nonlinear activated metasurface. Finally, the nonlinear activated metasurface constructed in this invention can achieve high-efficiency, nonlinear optical field manipulation in the reflection or transmission spectrum of optical fibers, exhibiting significant performance improvements compared to bare fibers, and can be applied in fields such as optical communication, optical protection, and visual enhancement. Therefore, this invention effectively overcomes the various shortcomings of existing technologies and has high industrial applicability.

[0126] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for fabricating a nonlinear activated metasurface integrated with an optical fiber endface, wherein the nonlinear activated metasurface comprises a plurality of spaced sub-metasurfaces, each sub-metasurface comprising a substrate and an array of nanopillars disposed on the substrate; the substrates of two adjacent sub-metasurfaces are connected by connecting pillars; characterized in that, The processing method includes the following steps: The fiber end face is pretreated to form an adhesion-promoting layer on the fiber end face; Uncured photoresist is coated onto the adhesion-promoting layer on the fiber end face; Construct a three-dimensional model of a nonlinear activated metasurface; Based on the three-dimensional model of the nonlinear activated metasurface, the photoresist on the fiber end face is subjected to two-photon polymerization exposure processing to form a nonlinear activated metasurface on the fiber end face. Gradient development and drying are performed to remove unexposed photoresist; The method for constructing the three-dimensional model of the nonlinear activated metasurface includes: An inverse neural network is used to solve the target output light field in reverse order to obtain the target phase matrix of each sub-metasurface in the nonlinear activated metasurface; the target phase matrix is ​​the phase delay φ(x,y) required by the nanopillars at different positions (x,y) on the substrate of the sub-metasurface. Based on the linear mapping relationship between column height and phase delay, the column height H(x,y) of each nanopillar in each sub-metasurface is calculated. Based on the coordinates and height of each nanopillar in each sub-metasurface, the preset substrate size, and the preset coordinates and size of the connecting pillars, a three-dimensional model of the nonlinear activated metasurface is generated using script-based automated modeling; the three-dimensional model is in STL format.

2. The processing method according to claim 1, characterized in that, Methods for obtaining the target phase matrix include: A forward physical propagation model is constructed using scalar diffraction theory or full-wave electromagnetic simulation algorithm; a randomly generated initial phase matrix is ​​input into the forward physical propagation model to calculate the corresponding light intensity distribution image; the light intensity distribution image obtained from the initial phase matrix is ​​converted into a wavefront amplitude intensity distribution to establish a phase matrix-wavefront amplitude intensity distribution dataset. An inverse neural network is constructed, and the wavefront amplitude intensity distribution is used as input and the phase matrix is ​​used as output. The constructed inverse neural network is trained by minimizing the loss function between the predicted phase matrix output by the inverse neural network and the actual phase matrix input to the inverse neural network. The weights of the inverse neural network are iteratively updated. The grayscale image of the target output light field is converted into the target wavefront amplitude intensity distribution, and the target wavefront amplitude intensity distribution is input into the trained inverse neural network to obtain the target phase matrix.

3. The processing method according to claim 1, characterized in that, The pretreatment of the optical fiber end face includes surface activation treatment and silane coupling agent grafting; the surface activation treatment includes oxygen plasma treatment and / or ultraviolet ozone treatment.

4. The processing method according to claim 1, characterized in that, The silane coupling agent used for grafting is 3-(trimethoxysilyl)propyl methacrylate.

5. The processing method according to claim 1, characterized in that, The pretreatment of the fiber end face also includes ultrasonic cleaning; the steps of ultrasonic cleaning are as follows: the fiber end face is ultrasonically cleaned for 5 minutes each in acetone, ethanol and ultrapure water, and then dried with nitrogen.

6. The processing method according to claim 1, characterized in that, The gradient development includes: The fiber end face with nonlinear activated metasurface was immersed in propylene glycol methyl ether acetate for 18 min. The fiber end face with the nonlinear activated metasurface was transferred and immersed in isopropanol for 2 to 3 minutes.

7. The processing method according to claim 1, characterized in that, After gradient development and drying, the fiber end face with nonlinear activated metasurface is cured at low temperature.

8. A nonlinear activated metasurface integrated at the end face of an optical fiber, characterized in that, The nonlinear activated metasurface is obtained by processing using the processing method described in any one of claims 1 to 7.

9. The nonlinear activated metasurface according to claim 1, characterized in that, The height of the nonlinear activated metasurface is 120 μm to 350 μm, and the interlayer spacing is 50 μm to 100 μm; the maximum lateral envelope diameter of the nanopillar array in the nonlinear activated metasurface is 15 μm to 250 μm.

10. An application of the nonlinear activated metasurface as described in claim 9 in environmental monitoring or biological detection.

Images