Integrated optical gas detection system based on multifunctional super surface and control method

Abstract

The application relates to the technical field of optical detection, in particular to an integrated optical gas detection system based on a multifunctional super surface and a control method, which comprises a super surface layer, a sensitive functional layer, a substrate and a photoelectric detector, the super surface layer is composed of a composite unit structure and is a single-layer / double-layer structure, a spatial light field reconstruction phase distribution is coded, the sensitive functional layer is composed of an optical adjustable material with reversible changes in optical properties with target environmental parameters, and the photoelectric detector is arranged at a preset spatial detection area. The super surface layer integrates sensitive response and light field regulation, uses optical property changes to modulate light intensity at a focal point, and converges incident light to the focal point through a phase distribution. Geometric parameters and spatial orientation angle distribution of the composite unit structure change optical response parameters with target environmental parameters, and a mapping relationship between an electrical signal and environmental parameters is established. The application can realize direct reading of environmental parameters only through the photoelectric detector, has high integration, low cost and fast response.

Description

Technical Field

[0001] This invention relates to the field of optical detection technology, and in particular to an integrated optical gas detection system and control method based on a multifunctional metasurface. Background Technology

[0002] Hydrogen is a clean energy source and an important industrial raw material, making its safe monitoring crucial. Optical hydrogen sensors have attracted widespread attention due to their inherent safety and resistance to electromagnetic interference. Existing optical hydrogen sensing schemes are mainly based on changes in the optical constants of palladium (Pd)-based materials, achieving hydrogen concentration sensing by measuring the transmittance, reflectance, or surface plasmon resonance (SPR) peak shift of the thin film.

[0003] However, traditional optical hydrogen sensing systems are typically complex and bulky. For example, SPR-based angle or wavelength spectrum measurement schemes require spectrometers or precision angle scanning mechanisms, resulting in low system integration and high costs. While metasurface hydrogen sensors, which have emerged in recent years, have achieved some miniaturization, their functions are relatively limited: either they only utilize the metasurface to achieve hydrogen-sensitive resonant absorption, requiring external spectral analysis equipment; or they use the metasurface as a phase modulation element, arranged separately from the hydrogen-sensitive thin film, failing to achieve true functional integration and simplified readout. In existing technologies, there is a complex spatial correspondence between the hydrogen-sensitive unit and the readout unit (such as detector arrays or imaging systems), resulting in long signal processing links and large data volumes, which limits their application in distributed, miniaturized, and low-power monitoring scenarios. Summary of the Invention

[0004] In view of this, the purpose of this invention is to propose an integrated optical gas detection system and control method based on a multifunctional metasurface, so as to solve the technical problems of existing optical hydrogen sensor systems being complex, having low integration, and having a lengthy readout link.

[0005] To achieve the above objectives, the present invention provides an integrated optical gas detection system based on a multifunctional metasurface, comprising: The metasurface layer is composed of multiple composite unit structures arranged in an array. The spatial distribution of the geometric parameters and / or in-plane orientation angles of the composite unit structures is configured to encode a preset spatial light field reconstructed phase distribution. The sensitive functional layer is composed of an optically tunable material whose optical properties change reversibly with the target environmental parameters, and is used to respond to changes in the target environmental parameters and modulate the light field; Substrate, used to support the metasurface layer; A photodetector is arranged in a preset spatial detection area of ​​the metasurface layer to convert the local optical signal in the preset spatial detection area into an electrical signal. The metasurface layer is a single-layer or double-layer structure. The sensitive functional layer is integrated with the metasurface layer. The metasurface layer modulates the light intensity at the preset spatial detection area by changing the optical properties of the optically tunable material, and focuses the incident light to the preset spatial detection area by reconstructing the phase distribution through the spatial light field. The geometric parameters and spatial orientation angle distribution of the composite unit structure are configured such that the optical response parameters of the system vary with the target environmental parameters, so as to establish a mapping relationship between the electrical signal and the target environmental parameters.

[0006] Preferably, the metasurface layer is a single-layer structure, the sensitive functional layer and the metasurface layer are composed of the same material layer, the composite unit structure encodes the spatial light field reconstructed phase distribution through its in-plane orientation angle distribution, and modulates the light intensity at the preset spatial detection area through the optical constant change caused by the target environmental parameters.

[0007] Preferably, the metasurface layer is a two-layer or multi-layer structure, comprising a sensitive functional layer, an isolation layer, and a wavefront modulation layer in sequence along the incident light direction, wherein: the sensitive functional layer is used to respond to target environmental parameters and modulate the incident light field; the intermediate isolation layer is used to isolate and support the sensitive functional layer and the wavefront modulation layer; and the wavefront modulation layer is used to form the spatial light field reconstruction phase distribution.

[0008] Preferably, the optically tunable material includes at least one of palladium, palladium alloy, palladium-based multilayer composite film, transition metal oxide, or phase change material, and the target environmental parameter is any one of hydrogen concentration, methane concentration, or nitric oxide concentration.

[0009] Preferably, the spatial light field reconstruction phase distribution includes an off-axis focusing phase component, which is configured to cause localized intensity enhancement of the incident light at the preset spatial detection region. The detection region may include, but is not limited to, a single point, multiple points, or a ring shape. The geometric parameters and / or in-plane orientation angle of the composite unit structure are calculated and determined based on the target phase distribution required for the preset spatial detection region. Preferably, the system operates in either a reflection or transmission mode, and the detected optical response parameters include any one of polarization conversion efficiency, transmittance, or reflectance.

[0010] Preferably, the geometric parameters and spatial distribution of the composite unit structure are determined by an inverse optimization algorithm. The inverse optimization algorithm takes the spatial light field reconstruction phase distribution and the optical contrast of the sensitive functional layer under different environmental parameters as optimization targets, and performs optimization in a preset structural parameter space. The inverse optimization algorithm evaluates the resolvability and monotonicity of the response of different structures under environmental parameter changes through a quality factor, constrains the monotonicity of the optical response parameters under different environmental parameter states and the separation degree between adjacent states, so that the optical response parameters change with the target environmental parameters.

[0011] Preferably, the inverse optimization algorithm includes an intelligent optimization strategy based on a surrogate model, wherein the surrogate model includes a deep neural network, Gaussian process regression, or random forest, and the global optimization mechanism of the intelligent optimization strategy includes at least one of Bayesian optimization framework, genetic algorithm, particle swarm optimization, topology optimization, gradient descent backpropagation, random search, or grid search.

[0012] Preferably, the reverse optimization algorithm includes an adaptive switching mechanism, which dynamically adjusts the exploration and development weights of the acquisition function according to the uncertainty or iterative convergence state of the surrogate model during the optimization process, so as to balance global search and local refinement.

[0013] This invention also provides a control method for an integrated optical gas detection system based on a multifunctional metasurface, applicable to any of the systems described above, comprising the following steps: After the incident light is modulated by the metasurface layer, the metasurface layer modulates the light intensity at the preset spatial detection area by changing the optical properties of the optically tunable material, and focuses the incident light to the preset spatial detection area by reconstructing the phase distribution through the spatial light field. The sensitive functional layer responds to changes in the target environmental parameters, and its optical properties undergo a reversible change, modulating the light intensity at the preset spatial detection area. The photodetector collects the optical signal at the preset spatial detection area and converts it into an electrical signal; The values ​​of the target environmental parameters are obtained by inversion based on the pre-calibrated correspondence between the electrical signal and the target environmental parameters.

[0014] The beneficial effects of this invention are: 1. This invention deeply integrates the sensitive response function and wavefront modulation function into a single-layer or double-layer metasurface, realizing an integrated "sensing-modulation" device design. Only an ordinary photodetector needs to be placed in the preset spatial detection area to complete the direct conversion from light intensity to electrical signal and then to environmental parameter value. It eliminates complex optical and electronic components such as spectrometers, charge-coupled device (CCD) arrays, angle scanning mechanisms or wavelength resolution elements, reducing the size, cost and power consumption of the sensing system, and improving the system robustness and response speed.

[0015] 2. This invention encompasses various metasurface structure schemes, including single-layer, double-layer, and multi-layer designs, offering significant design flexibility and application adaptability. In single-layer schemes, the sensitive functional layer and phase modulation layer are composed of the same material layer, resulting in a simple structure, simplified fabrication process, and high integration, making it suitable for applications sensitive to cost and fabrication complexity. In double-layer or multi-layer schemes, the sensitive functional layer and wavefront modulation layer are separated into different physical layers and optimized independently, enabling simultaneous high-sensitivity response and high-fidelity optical field modulation, suitable for applications requiring high detection accuracy and response speed. Different schemes can be flexibly selected according to specific application needs, providing differentiated technical paths for gas sensing in various scenarios.

[0016] 3. This invention encompasses two phase modulation methods: geometric phase modulation and transmission phase modulation, as well as two operating modes: transmission and reflection. The detected optical response parameters include polarization conversion efficiency, transmittance, or reflectance, which can be flexibly selected and switched according to specific application scenarios. By employing an inverse optimization algorithm, the precise design of the metasurface unit structure is achieved, establishing a mapping relationship between optical response parameters and target environmental parameters. This ensures that the system can invert environmental parameter values ​​through light intensity signals. Simultaneously, it integrates spatial light field reconstruction and spatial filtering functions, which helps suppress interference from non-target wavelength light.

[0017] 4. The metasurface structure of this invention can be mass-produced using mature micro-nano fabrication processes such as photolithography and etching, reducing the cost of large-scale production. The detection process eliminates the need for complex pretreatment procedures, shortening detection time, reducing equipment maintenance and manpower investment, and improving the economy and convenience of on-site industrial testing. The subwavelength scale and arrayable design of the metasurface make it naturally suitable for emerging fields such as multi-point distributed sensing, on-chip laboratories, and wearable monitoring devices, showing broad application prospects. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the Pd hydrogenation process in an embodiment of the present invention; (a) PdH x Schematic diagram of hydride formation; (b) Different PdH x Comparison of surface plasmon resonance absorption intensities under hydrogenated state parameters; Figure 2 The following are phase characteristic diagrams of Pd-based nanostructure units in embodiments of the present invention: (a) Curves showing the change of cross-polarized transmission phase with rotation angle under different Pd hydrogenation states; (b) Target phase distribution of off-axis superlens. Figure 3 This is a schematic diagram of the hydrogen-sensitive readout architecture integrated with a single-layer metasurface in an embodiment of the present invention; Figure 4 This is a flowchart of the adaptive switching neural Bayesian agent optimization algorithm in an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0021] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0022] This invention provides an integrated optical gas detection system based on a multifunctional metasurface. Along the optical path, it sequentially comprises a broadband incident light source, a multifunctional metasurface layer, and a substrate. A photodetector is positioned in a predetermined spatial detection area behind the metasurface layer. The broadband incident light source provides incident light covering the target's operating wavelength band. Its specific form can be a broadband halogen lamp, a supercontinuum light source, or a tunable laser, selected according to specific detection accuracy and integration requirements. The multifunctional metasurface layer consists of multiple composite unit structures arranged in an array, simultaneously serving as a sensitive response and wavefront modulation layer. The substrate is positioned below the metasurface layer to support it and provide mechanical strength. The substrate material is preferably a transparent dielectric material with low absorption loss in the operating wavelength band, such as quartz glass, silicon dioxide, or sapphire. The photodetector is positioned in the predetermined spatial detection area of ​​the metasurface layer to receive the local optical signal at that location and convert it into an electrical signal for output. Incident light enters from one side of the metasurface layer, and after being modulated by the metasurface layer, it forms a local light intensity maximum at a preset spatial detection area. The photosensitive surface of the photodetector is aligned with this focal position to achieve effective acquisition of the local light intensity.

[0023] Based on the above system architecture, the basic working principle of this invention is as follows: After incident light irradiates the metasurface layer, the metasurface layer modulates the light intensity at the preset spatial detection area by utilizing the optical property changes of the optically tunable material, and focuses the incident light to the preset spatial detection area by reconstructing the phase distribution through the spatial light field. The optical properties include, but are not limited to, complex permittivity, refractive index, extinction coefficient, absorption coefficient, transmittance, reflectivity, phase response, polarization response, or polarization conversion efficiency. When the target environmental parameters change, the optical constant of the sensitive functional layer undergoes a reversible change, which directly alters the modulation capability of the metasurface layer on the incident light, thereby changing the local light intensity at the preset spatial detection area. A photodetector positioned at this focal point collects the light intensity signal in real time and converts it into an electrical signal. Since the geometric parameters and spatial orientation angle distribution of the composite unit structure are configured to make the optical response parameters change with the target environmental parameters, a definite mapping relationship exists between the electrical signal and the target environmental parameters. The environmental parameters can be inverted through pre-calibration. The entire signal processing chain does not require complex optical and electronic components such as spectrometers, CCD arrays, or angle scanning mechanisms. It only requires a common photodetector to complete the direct conversion from light intensity to electrical signals and then to environmental parameter values.

[0024] It should be noted that a calibrable mapping relationship exists between the electrical signal and the target environmental parameters. In a preferred embodiment, the mapping relationship is a single-value mapping or a piecewise single-value mapping; when there are multi-value response intervals, disambiguation can be performed by combining preset detection intervals, historical change trends, auxiliary environmental parameters, or multi-point / multi-channel readout results, thereby determining the target environmental parameters. Its functional characteristics can be monotonic, non-monotonic, or even exhibit complex characteristics such as piecewise linearity under certain operating conditions. The embodiments of this application and the monotonic changes shown in the figures are merely one example, specifically illustrating the embodiments, and do not limit the scope of protection of this application.

[0025] The metasurface layer involved in this invention can be a single-layer structure, or a double-layer or multi-layer structure. Different structural schemes have different design ideas and advantages, and are suitable for different application scenarios. The single-layer structure and the double-layer structure are described in detail below with reference to the accompanying drawings. For the multi-layer structure, please refer to the double-layer structure, which will not be described again here.

[0026] Example 1: Single-layer metasurface structure Figure 3 This is a schematic diagram of an integrated optical hydrogen detection system with a single-layer metasurface structure, as described in an embodiment of the present invention. Figure 3 As shown, the system, along the optical path, sequentially includes a broadband incident light source, a multifunctional monolayer metasurface layer, and a substrate. A photodetector is positioned in a predetermined spatial detection area behind the metasurface layer. The broadband incident light source provides incident light covering the target's operating wavelength band. Its specific form can be a broadband halogen lamp, a supercontinuum light source, or a tunable laser, selected based on specific detection accuracy and integration requirements. The multifunctional monolayer metasurface layer consists of multiple composite unit structures arranged in an array, simultaneously serving as a sensitive response and wavefront modulation layer. The substrate is positioned below the metasurface layer to support it and provide mechanical strength. The substrate material is preferably a transparent dielectric material with low absorption loss in the operating wavelength band, such as quartz glass, silicon dioxide, or sapphire. The photodetector is positioned in the predetermined spatial detection area of ​​the metasurface layer to receive the local optical signal at that location and convert it into an electrical signal output. Incident light enters from one side of the metasurface layer, and after being modulated by the metasurface layer, it forms a local light intensity maximum at a preset spatial detection area. The photosensitive surface of the photodetector is aligned with this focal position to achieve effective acquisition of the local light intensity.

[0027] In this single-layer structure, the sensitive functional layer and the metasurface layer are composed of the same material layer. This material layer is made of an optically tunable material whose optical properties reversibly change with the target environmental parameters. Through the geometric configuration design of the unit structure, a plasmonic resonance response is generated in the target wavelength band. This single-layer material layer simultaneously undertakes the functions of sensitive response to environmental parameters and wavefront modulation of incident light. Specifically, the geometric parameters of the composite unit structure are designed so that its plasmonic resonance wavelength is located within the target operating wavelength band. When the environmental parameters change, the optical properties of the optically tunable material change, causing the plasmonic resonance condition of the unit structure to shift, thereby changing its optical response parameters. At the same time, by assigning different in-plane orientation angles to the unit structures at different spatial locations, the phase distribution is reconstructed by encoding the preset spatial light field using the geometric phase principle. A nonlinear mapping relationship is pre-established between the orientation angle of the composite unit structure and the cross-polarized transmission phase through full-wave electromagnetic simulation. The required orientation angle for each spatial location is deduced based on the preset target phase distribution, thus achieving both sensitive response and wavefront modulation functions in a single material layer. The composite unit structure can be one or more of nanorods, nanopillars, nanorings, or nanocross structures, and of course, it can also be other nano-shapes, which are not limited here.

[0028] Taking Pd-based materials for hydrogen detection as an example, Figure 1 The principle of the Pd hydrogenation process is shown. Figure 1 (a) is PdH x Schematic diagram of hydride formation. When Pd comes into contact with hydrogen gas, hydrogen molecules dissociate into hydrogen atoms on the Pd surface, and these hydrogen atoms diffuse into the Pd crystal lattice to form PdH. x The hydride phase. As the hydride state parameter x increases from 0 to 0.5, the complex permittivity of Pd changes significantly. Figure 1 (b) for different PdH x Comparison of plasmonic resonance absorption intensities under hydride state parameters. The figure clearly shows that as the hydride state parameters gradually increase, the intensity of the resonance absorption peak changes significantly. This directly reflects the change in polarization conversion efficiency, providing a physical basis for the optical detection of hydrogen concentration. Figure 1 All coordinate axes are labeled in Chinese for easy and intuitive understanding.

[0029] In one specific implementation of this embodiment, the composite unit structure uses Pd nanorods with the following geometric parameters: height H = 329 nm, period P = 689 nm, length L = 507 nm, and width W = 304 nm. The substrate material is silicon dioxide (SiO2). These structural parameters are determined using an inverse optimization algorithm, ensuring that the polarization conversion efficiency varies with the hydride state parameter x within the target operating wavelength band, and that the polarization conversion efficiency between each hydride state is sufficiently distinguishable. The geometric parameters of the Pd nanorods are designed to generate plasmon resonance in the target wavelength band, for example, from 690 nm to 770 nm. When Pd contacts hydrogen gas, PdH forms. x After hydride phase, its complex permittivity changes, the plasmon resonance absorption characteristics change accordingly, and the polarization conversion efficiency undergoes a detectable shift.

[0030] Regarding the implementation method of spatial light field reconstruction, Figure 2 The phase characteristics of Pd-based nanostructure units are shown. Figure 2 (a) shows the curves of cross-polarized transmission phase versus rotation angle under different Pd hydride states. The cross-polarized transmission phase of conventional geometric phase metasurfaces exhibits a linear relationship φ with the unit cell orientation angle. cross =±2σθ, where σ=±1 corresponds to left-handed and right-handed circularly polarized incident light, respectively. However, the Pd-based nanostructure used in this invention exhibits a significant plasmon resonance response near 730 nm, and φ under different hydrogenation states... cross The phase changes nonlinearly with θ. Therefore, in this embodiment, full-wave electromagnetic simulations such as the finite-difference time-domain (FDTD) method are used to pre-scan the cross-polarized transmission phase φ at each rotation angle from 0° to 180°. cross A nonlinear mapping fitting relationship between the rotation angle and the phase is established. Figure 2 (b) is a schematic diagram of the target phase distribution of the off-axis superlens.

[0031] The spatial light field reconstruction phase distribution includes an off-axis focusing phase component, which is configured to enhance the local light intensity of the incident light at the preset spatial detection region. The preset spatial detection region includes, but is not limited to, a single point, multiple points, or a ring shape, or even other irregular shapes. The geometric parameters and / or in-plane orientation angles of the composite unit structure are calculated and determined according to the target phase distribution required by the preset spatial detection region. The calculation method can be determined by nonlinear fitting relationship or direct calculation and back deduction, or other calculation methods may also exist.

[0032] For example, when the preset spatial detection area is a single point, the off-axis focusing phase is calculated as follows: ; in To design the center wavelength, ( , , () represents the preset focus coordinates. These are the spatial coordinates of the composite unit structure on the metasurface layer. This represents the target phase at that location.

[0033] Based on the preset off-axis focusing parameters, the target phase φ required for each spatial position (x, y) is calculated according to the target phase distribution formula. desired (x,y), and then the orientation angle θ(x,y) that should be set at this position is deduced by the aforementioned fitting relationship. In this way, while keeping the geometry of the nanorod unchanged, the off-axis focused wavefront can be encoded simply by the distribution of the spatial orientation angle.

[0034] Example 2: Double-layer metasurface structure In this embodiment, the metasurface layer has a two-layer structure, comprising an isolation layer and a wavefront modulation layer. Adding a sensing functional layer integrated with the metasurface layer, the structure along the incident light direction sequentially includes a sensing functional layer, an isolation layer, and a wavefront modulation layer. The sensing functional layer is used to respond to target environmental parameters and modulate the incident light field. The intermediate isolation layer isolates and supports the sensing functional layer and the wavefront modulation layer. The wavefront modulation layer is used to form the reconstructed phase distribution of the spatial light field.

[0035] If the metasurface layer is a multi-layer structure, that is, in addition to the isolation layer and the wavefront modulation layer, other layers can be set, plus the sensitive functional layer integrated with the metasurface layer, together forming the complete optical system that this application intends to protect. For specific implementation methods, please refer to the above-mentioned double-layer structure, which will not be repeated here.

[0036] Compared to single-layer structures, dual-layer or multi-layer structures separate the sensing function and the optical field manipulation function into different physical layers, each optimized independently, enabling simultaneous high-sensitivity response and high-fidelity optical field manipulation. In a dual-layer architecture, incident light sequentially passes through the sensing layer, the isolation layer, and the wavefront manipulation layer along the optical path, finally converging at the photodetector in the predetermined spatial detection area. The spatial relative positions of each layer and the optical path orientation constitute the complete optical system of the dual-layer or multi-layer structure.

[0037] The selection of materials and thickness design of the sensitive functional layer are primarily optimized for environmental sensitivity response performance. It can be implemented as a continuous thin film or a patterned array structure. Materials for the sensitive functional layer may include at least one of palladium, palladium alloys, palladium-based multilayer films, transition metal oxides, two-dimensional materials, polymer sensitive materials, or composite materials thereof. When the sensitive material comes into contact with the target environment, such as hydrogen, its optical properties reversibly change with environmental parameters such as the degree of hydrogenation, thereby altering the transmittance or reflectance of incident light and applying phase modulation to the transmitted or reflected light field that varies with environmental parameters. These optical properties include, but are not limited to, complex permittivity, refractive index, extinction coefficient, absorption coefficient, transmittance, reflectance, phase response, polarization response, or polarization conversion efficiency.

[0038] An isolation layer is positioned between the sensitive functional layer and the wavefront modulation layer. It serves two purposes: firstly, it provides physical isolation, preventing mechanical contact and mutual interference between the two layers; secondly, it provides optical isolation, reducing optical crosstalk between the upper and lower layers by selecting appropriate isolation layer materials and thicknesses. The isolation layer material can include at least one of silicon dioxide, silicon nitride, alumina, or a polymer dielectric, preferably a low-refractive-index dielectric material such as SiO2, which offers good isolation performance and is easily fabricated using conventional thin-film deposition processes. The thickness of the isolation layer is determined based on the operating wavelength and the specific design of the upper and lower layer structures, typically on the order of hundreds of nanometers.

[0039] The wavefront modulation layer is composed of a high-refractive-index dielectric material, which can be selected from at least one of silicon (Si), titanium dioxide (TiO2), gallium nitride (GaN), silicon nitride (Si3N4), or other high-refractive-index dielectric materials. These materials have high refractive indices and low absorption losses in the visible to near-infrared band, making them suitable for constructing high-efficiency metasurface phase modulation elements. The composite unit structure of the lower layer can adopt isotropic or anisotropic geometries such as nanopillars or nanofins, depending on the selected phase modulation method.

[0040] In this embodiment, the lower wavefront modulation layer can reconstruct the spatial optical field of the incident light through geometric phase modulation or transmission phase modulation.

[0041] For the geometric phase modulation method, each composite unit structure in the lower layer has an in-plane anisotropic geometry and a preset orientation angle. By assigning different orientation angles to different spatial positions, precise control of the wavefront phase is achieved while keeping the unit geometric parameters constant. Similar to the orientation angle encoding of Pd nanorods in a single-layer structure, each nanostructure unit of the lower Si superlens in a bilayer structure has a preset orientation angle θ. Incident circularly polarized light undergoes polarization conversion after passing through the anisotropic nanostructure, and the cross-polarization components obtain a geometric phase φ = ±2σθ, where σ = ±1 corresponds to left-handed and right-handed circularly polarized incident light, respectively. The preset target phase distribution is encoded by the spatially distributed orientation angles. The advantages of the geometric phase method are low dispersion, intuitive design, and consistent geometric parameters of the unit structure, such as length, width, and height, during manufacturing. Only the orientation angle varies with spatial position, reducing manufacturing complexity.

[0042] For the transmission phase modulation (TPM) method, each composite unit structure in the lower layer has an in-plane isotropic or anisotropic geometry. By changing the geometric parameters of the composite unit structure at different spatial locations, including length, width, height, and period, the equivalent refractive index of the unit structure is altered to achieve precise control of the wavefront phase. The physical mechanism of TPM is as follows: when the characteristic size of the unit structure is much smaller than the operating wavelength, the metasurface layer can be considered as a homogeneous dielectric layer with an equivalent refractive index, which varies with the duty cycle, height, and morphological parameters of the unit structure. By designing the unit structure size at different spatial locations, different phase delays are accumulated as the incident light passes through the metasurface layer, thus forming a predetermined spatial phase distribution. The advantage of the TPM method is its large phase coverage, covering the entire range from 0 to 2π, and its insensitivity to the polarization state of the incident light, making it suitable for unpolarized light incident scenarios.

[0043] In the specific optical path of the dual-layer structure, the incident light first passes through the upper sensitive functional layer, which modulates the amplitude and / or phase of the incident light according to the target environmental parameters. Subsequently, it propagates through the intermediate isolation layer to the lower wavefront modulation layer. The lower wavefront modulation layer applies a preset spatial optical field to the incident light to reconstruct the phase distribution, converging the optical field carrying the target environmental parameter information to a preset spatial detection area. A photodetector is placed at the off-axis focal point to collect the local light intensity and convert it into an electrical signal. Because the modulation of the optical field by the upper sensitive functional layer varies with the target environmental parameters, the final light intensity at the focal point also changes accordingly. The environmental parameters can then be inverted using a pre-calibrated curve showing the relationship between the electrical signal and the target environmental parameters.

[0044] A key advantage of the dual-layer structure is that the upper sensing layer and the lower wavefront modulation layer can be independently optimized. The thickness of the upper sensing layer can be independently adjusted according to the required response speed and modulation depth, without being limited by the structural parameters of the lower superlens. For example, a thinner Pd film has a faster hydrogen adsorption-desorption kinetic response speed but a smaller modulation depth, while a thicker Pd film has a larger modulation depth but a slower response speed. The optimal balance between the two can be achieved by independently adjusting the thickness of the upper film according to the specific application scenario, without affecting the focusing performance of the lower superlens. The structural parameters of the lower superlens can be independently optimized according to the required focusing efficiency and imaging quality, without being limited by the optical properties of the upper sensing material. The two are physically and optically isolated by an intermediate isolation layer, avoiding mutual interference. In contrast, in a single-layer structure, because the sensing material and the phase modulation material are in the same material layer, all properties are coupled together, and the adjustment margin is limited. In applications that require both high response speed and detection accuracy, the dual-layer structure has a significant advantage. This dual-layer structure is also applicable to both transmission and reflection modes. In the reflection mode, a continuous metal reflection layer can be added below the lower phase modulation layer, and the incident light and the emitted light are located on the same side of the metasurface chip, which facilitates system integration and packaging.

[0045] Regarding the operating modes, the system of this invention can employ either a transmission or a reflection mode. In the transmission mode, incident light enters from one side of the metasurface chip and exits from the other side, with a photodetector positioned at a predetermined spatial detection area on the exiting side of the metasurface chip. In the reflection mode, a continuous metal reflective layer is disposed below the lower phase modulation layer or the substrate, and incident light enters from the front side of the metasurface chip, exits from the front after reflection, with a photodetector positioned at a predetermined spatial detection area on the incident side of the metasurface chip. The two operating modes can be flexibly selected based on the spatial layout, installation conditions, and detection accuracy requirements of the specific application scenario. The system's optical response parameters include any one of polarization conversion efficiency, transmittance, or reflectance. When detecting polarization conversion efficiency, the incident light is circularly polarized, and the cross-polarization component is detected at the exit end; when detecting transmittance or reflectance, the incident light is linearly polarized or unpolarized, and the intensity change of the transmitted or reflected light is directly detected.

[0046] In this embodiment, the incident light source is circularly polarized light, preferably left-handed circularly polarized light. When the incident light irradiates the metasurface layer, each composite unit structure applies spatially varying phase modulation to the incident light according to a preset orientation angle distribution, causing the cross-polarized components to interfere and enhance at a preset spatial detection region, forming a local intensity maxima. When the hydrogenation state of Pd changes, the plasmon resonance absorption characteristics of the unit structure change, leading to a change in the cross-polarization conversion efficiency, and the light intensity at the focal point changes accordingly. The photodetector collects the local light intensity at this focal point and converts it into photocurrent. By using a pre-calibrated relationship curve between the photocurrent and the hydrogen concentration, the concentration can be inverted in real time.

[0047] Regarding reversibility, the Pd hydrogenation process undergoes a reverse dehydrogenation reaction when the hydrogen concentration decreases, resulting in PdH2. x The hydrogenation reaction decomposes into Pd and hydrogen, restoring the dielectric constant to its initial state and the focal light intensity accordingly, thus achieving reversible hydrogen concentration detection. The operating temperature affects both the equilibrium constant and the response rate of the Pd hydrogenation reaction; temperature compensation can be implemented by integrating a temperature sensor into the system to improve detection accuracy.

[0048] When the hydrogen concentration increases, hydrogen atoms enter the octahedral interstitial positions of the Pd lattice, forming a solid solution of α phase (low hydrogen concentration) and β phase (high hydrogen concentration). The lattice constant of Pd increases with the degree of hydrogenation x, and the complex dielectric constant changes accordingly. When the hydrogen concentration decreases, hydrogen atoms are extracted from the Pd lattice, the lattice constant recovers, and the dielectric constant returns to its initial state. The response time of this reversible process is related to factors such as the thickness of the Pd film, grain size, operating temperature, and hydrogen concentration gradient. In a preferred embodiment, by integrating a temperature sensor into the system, multiple concentration-electrical signal response curves are calibrated at different ambient temperatures. The signal processing unit selects the appropriate calibration curve or performs interpolation calculations based on the real-time readings of the temperature sensor, thereby eliminating the influence of temperature on the detection results.

[0049] Example 3: Inverse Optimization Algorithm To achieve precise design of the geometric parameters of the composite unit structure and ensure that the optical response parameters vary with the target environmental parameters and that there is sufficient resolution between each response state, this invention employs an inverse optimization algorithm to determine the geometric parameters and spatial distribution of the composite unit structure.

[0050] Figure 4 A flowchart of the adaptive switching neural-Bayesian agent optimization algorithm in an embodiment of the present invention is shown. The core of the inverse optimization algorithm is to optimize the target phase distribution and the optical contrast of the sensitive functional layer under different environmental parameters within a preset structural parameter space. The inverse optimization algorithm does not depend on a specific optimization algorithm, but rather encompasses multiple numerical optimization methods that can achieve the same optimization objective.

[0051] Taking Pd-based nanorods for hydrogen detection as an example, the optimization variables include the period P, width W, length L, and height H of the unit structure. For palladium-based nanorod structures, the optimization variables range as follows: period P is set to 500 nm to 800 nm, width W to 200 nm to 400 nm, length L to 400 nm to 600 nm, and height H to 200 nm to 500 nm. The optimization objective is to maximize the polarization conversion efficiency as the Pd hydride state parameter x∈{0,0.125, 0.25, 0.5} varies with the target operating wavelength range, such as 690 nm to 770 nm, while also maximizing the separation of polarization conversion efficiency between adjacent hydride states. Optical response parameters, including but not limited to polarization conversion efficiency or transmittance, vary with the target environmental parameters, which is a core prerequisite for ensuring that the single-point readout system can uniquely invert the environmental parameter values ​​through the light intensity signal.

[0052] The inverse optimization algorithm uses a surrogate model-assisted intelligent optimization strategy for global optimization. The surrogate model replaces computationally expensive full-wave electromagnetic simulation, accelerating the optimization iteration process. Surrogate models include, but are not limited to, deep neural networks, Gaussian process regression, or random forests. Deep neural networks have the advantages of strong fitting ability and suitability for high-dimensional parameter spaces; Gaussian process regression has a natural uncertainty estimation capability, suitable for Bayesian optimization frameworks; random forests have the advantages of fast training speed and good robustness to outliers. Global optimization mechanisms include, but are not limited to, any one or a combination of Bayesian optimization frameworks, genetic algorithms, particle swarm optimization, topology optimization, gradient descent backpropagation, random search, or grid search. Bayesian optimization frameworks balance exploration and development through a collection function, approximating the global optimum with fewer iterations; genetic algorithms and particle swarm optimization are swarm intelligence algorithms, suitable for multi-peak optimization problems; topology optimization can freely search for arbitrary geometric configurations within the design space; gradient descent backpropagation utilizes gradient information for rapid convergence; random search and grid search are simple to implement and suitable as baseline comparison methods.

[0053] Preferably, the inverse optimization algorithm includes an adaptive switching mechanism. During the optimization process, the exploration and development weights of the acquisition function are dynamically adjusted based on the uncertainty or iterative convergence state of the current surrogate model to balance global search and local refinement. Specifically, in the early stages of optimization, when the surrogate model has lower accuracy, the acquisition function focuses on exploration, i.e., it prioritizes sampling areas with high model uncertainty to quickly establish an understanding of the global design space. In the later stages of optimization, when the surrogate model has improved accuracy, the acquisition function focuses on development, i.e., it prioritizes dense sampling of areas where the model predicts the best solution to refine local optima. The adaptive switching mechanism automatically determines whether the current stage is early or late by monitoring the prediction variance or the slope of the iterative convergence curve of the surrogate model, and dynamically adjusts the hyperparameters of the acquisition function, achieving efficient global optimization without manual intervention.

[0054] In the specific implementation of the inverse optimization algorithm, an initial sample set is first generated in the preset structural parameter space using the Latin hypercube sampling method. The number of samples is typically 10-20 times the parameter dimension (e.g., for 4 optimization variables, the initial sample size is 40-80 sets). Each sample set corresponds to a specific set of Pd nanorod geometric parameters (P, W, L, H), and its polarization conversion efficiency spectrum in the target working band is calculated using full-wave electromagnetic simulation (such as FDTD). Subsequently, a deep neural network surrogate model is trained using these simulation data, with the structural parameters as input and the polarization conversion efficiency spectrum as output. After the surrogate model is trained, a Bayesian optimization framework is used for global optimization: in each iteration step, the trained surrogate model is used to predict the optical response and uncertainty of candidate points, and exploration and development are balanced by a collection function (such as the expected improvement function). The candidate point with the largest collection function value is selected for high-precision FDTD simulation verification, and the verification results are fed back to the training set to update the surrogate model. The above iterative process is repeated until the convergence condition is met (e.g., the maximum number of iterations reaches a preset value, or the optimal quality factor no longer improves after several consecutive iterations). The adaptive switching mechanism automatically adjusts the weight parameters of the acquisition function based on the uncertainty of the proxy model, focusing on exploration in the early stage of optimization (when uncertainty is high) and on development in the later stage of optimization (when uncertainty is low).

[0055] To quantitatively evaluate the distinguishability and monotonicity of the responses of different structures under varying environmental parameters, a quality factor (FoM) is defined. Let four representative response state parameters be x∈{0, 0.125, 0.25, 0.5}, and let λ be a sample wavelength point within the target operating band. j The optical response parameter measured at point T is denoted as T. i,j Where subscript i corresponds to the i-th hydrogenation state, and subscript j corresponds to the j-th sampling wavelength. The separation of optical response parameters between adjacent hydrogenation states is defined as d. k,j = |T k+1,j T k,j | where k = 1, 2, 3 corresponds to three pairs of adjacent states, namely 0 and 0.125, 0.125 and 0.25, and 0.25 and 0.5, and j traverses all sampling wavelength points. For each pair of adjacent states k, its maximum separation is denoted as D. kThe global minimum interval is denoted as g. To ensure that the sensing response can be inverted in a single value, the optical response parameters must vary with environmental parameters at all sampling wavelengths. A monotonicity factor M is defined: if all wavelengths satisfy strict monotonicity, then M = +1; if any wavelength does not satisfy strict monotonicity, then M = -1. The definition of FoM constrains the response monotonicity (M = +1) and the minimum separation between adjacent states (g > 0), while limiting the wavelength position corresponding to the maximum separation to the middle of the optimization interval, for example, within the range of 700 nm to 800 nm, to ensure that the central operating frequency is located in this region, thus balancing detection sensitivity and operating bandwidth. A higher FoM helps to obtain a more stable monotonic response and better response state resolvability. The specific definition of FoM can be adjusted according to different application requirements and optimization objectives, as long as its core function is to evaluate monotonicity and resolvability.

[0056] The definition of the quality factor FoM above is only an example. In the actual optimization process, the specific mathematical form of FoM can be adjusted and modified according to the range of target environmental parameters, working band width and specific requirements of detection accuracy. However, the basic purpose is to achieve a quantitative evaluation of the monotonicity of optical response parameters and the distinguishability between adjacent states.

[0057] Example 4: Single-point readout architecture Based on the aforementioned single-layer or double-layer metasurface structure, this embodiment designs the metasurface to converge incident light to a preset spatial detection region. The light intensity at the preset spatial detection region is directly modulated by the change in the optical constant of the sensitive functional layer. At the same time, off-axis focusing itself realizes a spatial filtering function, that is, only light components that meet specific phase matching conditions can undergo interference enhancement at the focal point, while light components in other directions are spatially separated and filtered out.

[0058] A photodetector is positioned at a predetermined spatial detection area to receive local optical signals at that location and convert them into electrical signals. The photodetector is selected from one of the following: a Si-PIN photodiode, avalanche photodiode, or photomultiplier tube. Si-PIN photodiodes offer advantages such as low cost, low dark current, and a wide linear range, making them suitable for detection scenarios with standard sensitivity requirements. Avalanche photodiodes have internal gain and high sensitivity, making them suitable for low-light detection scenarios. Photomultiplier tubes offer extremely high sensitivity and time resolution, making them suitable for ultra-weak light signals and fast-response scenarios. The detector selection is determined based on the incident light power, target detection sensitivity, and response speed requirements.

[0059] The electrical signal output by the photodetector is a photocurrent or photovoltage, the amplitude of which is proportional to the local light intensity at the preset spatial detection area. Because the composite unit structure of the metasurface layer is designed to allow optical response parameters to vary with target environmental parameters, a one-to-one mapping relationship exists between the electrical signal and the target environmental parameters. By pre-calibrating and establishing the correspondence curve between the electrical signal amplitude and the target environmental parameters, in subsequent detection processes, only the output electrical signal of the photodetector needs to be read to obtain the current value of the target environmental parameters from the calibration curve. The entire signal processing chain requires no spectrometer, CCD array, angle scanning mechanism, or wavelength resolution element; the direct conversion from optical signal to environmental parameter value can be completed solely through the output of a single photodetector.

[0060] Regarding the dynamic response characteristics of the signal, when the target environmental parameters undergo a step change, the optical constant of the sensitive functional layer changes exponentially with the adsorption or desorption process of the sensitive material. The light intensity at the off-axis focal point responds dynamically accordingly, and the output electrical signal of the photodetector exhibits a corresponding transient change. By monitoring the rate of change of the electrical signal over time, dynamic change information of the target environmental parameters can also be obtained, such as the rate of increase or decrease in hydrogen concentration. This has additional value for certain applications requiring early warning, such as hydrogen leak monitoring.

[0061] Another advantage of the spatially filtered single-point readout architecture is its natural suppression of interference from non-target wavelength light. Off-axis focusing phase components are wavelength-dependent; different wavelengths of light have different focusing positions after passing through the metasurface layer. Therefore, the target wavelength light component can only achieve interference enhancement at a predetermined spatial detection region corresponding to the target operating wavelength; non-target wavelength light components cannot satisfy the constructive interference condition at this location and thus will not cause significant interference to the signal. This spatial filtering characteristic eliminates the need for additional filters or spectrometers, achieving this directly at the metasurface level, further simplifying the system structure.

[0062] Regarding the manufacturing method, the metasurface structure of this invention can be patterned using electron beam lithography, nanoimprint lithography, or deep ultraviolet lithography, and then the pattern can be transferred to the material layer by reactive ion etching or inductively coupled plasma etching to form an array of nanopillars or nanorods with a high aspect ratio. The fabrication of the bilayer structure requires sequential deposition and patterning of each material layer. For a single-layer structure, the process flow is as follows: deposit an optically tunable material thin film on a substrate, spin-coat photoresist, define the array pattern using photolithography or electron beam exposure, etch to form a composite unit structure, and remove residual photoresist to obtain the finished product. For a bilayer structure, the process flow is as follows: deposit a lower high-refractive-index dielectric material on a substrate and pattern it to form a phase modulation layer, then deposit an intermediate isolation layer, and finally deposit an upper sensitive material layer on the isolation layer and pattern it as needed. The above fabrication methods are all mature processes in the field of semiconductor micro / nano fabrication and are suitable for mass production.

[0063] In summary, the integrated optical gas detection system based on a multifunctional metasurface proposed in this invention offers both single-layer and double-layer structural options. It employs geometric phase modulation or transmission phase modulation to reconstruct the spatial light field and can operate in either transmission or reflection modes. It detects optical response parameters such as polarization conversion efficiency or transmittance. An inverse optimization algorithm ensures the relationship between optical response parameters and target environmental parameters. Finally, a photodetector positioned in a predetermined spatial detection area directly converts light intensity into an electrical signal and then into environmental parameter values. Simulation results demonstrate that this readout architecture effectively converts environmental parameter information into electrical signals, and the electrical signals exhibit a good monotonic correspondence and repeatability with the target environmental parameters. This invention integrates sensitive response, wavefront modulation, and single-point readout into a single-layer or double-layer metasurface platform, simplifying the complex readout chain of traditional optical sensing systems. It provides a complete technical solution for high-performance, miniaturized, and low-cost environmental parameter detection, showing promising application prospects in distributed gas sensing, on-chip laboratories, and wearable monitoring devices.

[0064] Example 5: Control method for an integrated optical hydrogen detection system This embodiment provides a control method for an integrated optical gas detection system based on a multifunctional metasurface, building upon the aforementioned single-layer or double-layer metasurface structures. This method is suitable for real-time detection of target environmental parameters such as hydrogen concentration, methane concentration, or nitric oxide concentration, and is particularly applicable to applications with strict requirements on size and power consumption, such as distributed gas sensing, on-chip laboratories, and wearable monitoring devices.

[0065] First, based on the detection range and accuracy requirements of the target environmental parameters, a single-layer or double-layer metasurface structure is selected, and corresponding composite unit structure parameters are configured. For applications requiring rapid response and high detection accuracy, a double-layer metasurface structure is preferred. The upper sensitive functional layer is composed of materials including, but not limited to, pure palladium or palladium alloys, while the lower wavefront modulation layer uses a high-refractive-index dielectric material superlens. The two are physically and optically isolated by an intermediate isolation layer. For applications sensitive to cost and fabrication complexity, a single-layer metasurface structure is preferred. The sensitive functional layer and the wavefront modulation layer are composed of the same material layer, such as a Pd nanorod array. Sensitive response and off-axis focusing wavefront encoding are simultaneously achieved through spatial orientation angle distribution. The design center wavelength and off-axis focus coordinates of the metasurface layer are determined based on the characteristics of the incident light source and the system spatial layout. For example, the design center wavelength is 730 nm, and the off-axis focus coordinates are (…). =30 μm, =0, =50 μm), aperture radius, for example, 30 μm. After the metasurface layer is fabricated, it is integrated with the substrate, broadband incident light source and photodetector according to the aforementioned single-layer or double-layer structure architecture to ensure that the photosensitive surface of the photodetector is precisely aligned with the preset spatial detection area.

[0066] During the signal acquisition phase after system startup, the incident light is modulated by the metasurface layer. The metasurface layer utilizes the optical properties of the optically tunable material to modulate the light intensity at the preset spatial detection area, and reconstructs the phase distribution through the spatial light field to focus the incident light onto the preset spatial detection area. Specifically, a broadband incident light source emits a beam covering the target's operating wavelength band, and the beam illuminates the metasurface layer with a preset polarization state (circularly polarized or linearly polarized light). When the incident light is circularly polarized, the composite unit structure of the metasurface layer encodes the off-axis focusing phase distribution through geometric phase encoding, focusing the cross-polarized components onto the preset off-axis focus; when the incident light is linearly polarized, the metasurface layer encodes the off-axis focusing phase distribution through transmission phase encoding, focusing the transmitted or reflected light onto the preset off-axis focus. Changes in the target environmental parameters cause a reversible change in the dielectric constant of the sensitive functional layer. This change directly alters the phase modulation or amplitude modulation of the incident light by the metasurface layer, thereby causing a corresponding change in the local light intensity at the preset spatial detection area.

[0067] Subsequently, in response to changes in the target environmental parameters, the dielectric constant of the sensitive functional layer undergoes a reversible change, modulating the light intensity at the preset spatial detection area. Specifically, taking Pd-based materials for hydrogen detection as an example, when hydrogen molecules in the environment come into contact with the surface of the Pd-based sensitive material, the hydrogen molecules catalytically dissociate into hydrogen atoms on the Pd surface, and the hydrogen atoms diffuse into the Pd lattice to form PdH. x In the hydride phase, the complex permittivity of Pd changes with the hydride state parameter x. As the hydride state parameter x increases, the plasmon resonance absorption characteristics of Pd shift, manifested as changes in polarization conversion efficiency or transmittance. When the hydrogen concentration decreases, PdH... x A reverse dehydrogenation reaction occurs, and hydrogen atoms in the Pd lattice are released as hydrogen gas, restoring the dielectric constant of Pd to its initial state, thus enabling reversible detection.

[0068] Then, the photodetector acquires the optical signal at a preset spatial detection area and converts it into an electrical signal. The photodetector is selected from Si-PIN photodiodes, avalanche photodiodes, or photomultiplier tubes, and its output signal is photocurrent or photovoltage. The amplitude of this electrical signal is proportional to the local light intensity at the preset spatial detection area, and the local light intensity varies with the dielectric constant of the sensitive functional layer. Therefore, the amplitude of the electrical signal indirectly reflects the current value of the target environmental parameters. The output signal of the photodetector is processed by a preamplifier and an analog-to-digital converter before being input to the signal processing unit. Since there is only one photodetector channel in the entire system, the signal processing link is extremely simple, avoiding complex problems such as multi-channel signal synchronization, data splicing, and inter-channel calibration.

[0069] In the signal processing unit, the values ​​of the target environmental parameters are obtained by inversion based on the pre-calibrated correspondence between the electrical signal and the target environmental parameters. The pre-calibration process is as follows: before the system is put into use, it is placed in a target gas environment with several known concentrations, and the amplitude of the output electrical signal of the photodetector corresponding to each concentration is recorded to establish an electrical signal amplitude-gas concentration calibration curve. During the calibration process, since the composite unit structure of the metasurface layer is designed to make the optical response parameters change with the target environmental parameters, each electrical signal amplitude uniquely corresponds to one environmental parameter value, and there is no ambiguity of multiple values. During the actual detection process, the signal processing unit compares the real-time acquired electrical signal amplitude with the calibration curve and calculates the current environmental parameter value by looking up a table or interpolation. For hydrogen concentration detection, the horizontal axis of the calibration curve is the hydrogen concentration value, and the vertical axis is the output electrical signal amplitude of the photodetector, and the two have a monotonic one-to-one correspondence. For applications requiring temperature compensation, the calibration process is carried out separately at different ambient temperatures to establish multiple temperature-calibration curves. The signal processing unit selects the corresponding calibration curve for inversion calculation based on the real-time readings of the temperature sensor integrated in the system. For other target gases such as methane or nitric oxide, the basic principle of the calibration curve is the same, only the material selection of the sensitive functional layer and the type of calibration gas are adjusted accordingly.

[0070] During dynamic monitoring, when the target environmental parameters undergo a step change or continuous change, the above four steps are executed in real time in a loop. The photodetector continuously outputs an electrical signal reflecting the current environmental parameters, and the signal processing unit reads this electrical signal at a preset sampling frequency and updates the displayed or output environmental parameter values ​​in real time. By monitoring the rate of change of the electrical signal over time, dynamic change information of the target environmental parameters can also be obtained, such as the rate of increase or decrease in hydrogen concentration. This is of great value for applications requiring early warning (such as hydrogen leak monitoring).

[0071] Throughout the entire control process, the signal processing link requires no spectrometer, CCD array, angle scanning mechanism, or wavelength resolution element; it directly converts optical signals into environmental parameter values ​​using only the output of a photodetector. Spatial filtering is achieved by the off-axis focusing phase distribution itself: only light components matching the target operating wavelength meet the preset phase matching condition, resulting in enhanced interference at the off-axis focal point. Light components of non-target wavelengths cannot form constructive interference at this location, thus avoiding significant interference to the measurement signal. This spatial filtering characteristic eliminates the need for additional filters or wavelength selection devices, further simplifying the system structure and reducing manufacturing costs and system power consumption.

[0072] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0073] In the embodiments provided in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0074] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0075] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0076] The implementation of all or part of the processes in the methods of the above embodiments can also be accomplished by a computer program product. When the computer program product is run on a terminal device, the terminal device can implement the steps in the various method embodiments described above.

[0077] The embodiments described above are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An integrated optical gas detection system based on a multifunctional metasurface, characterized in that, include: The metasurface layer is composed of multiple composite unit structures arranged in an array. The spatial distribution of the geometric parameters and / or in-plane orientation angles of the composite unit structures is configured to encode a preset spatial light field reconstructed phase distribution. The sensitive functional layer is composed of an optically tunable material whose optical properties change reversibly with the target environmental parameters, and is used to respond to changes in the target environmental parameters and modulate the light field; Substrate, used to support the metasurface layer; A photodetector is arranged in a preset spatial detection area of ​​the metasurface layer to convert the local optical signal in the preset spatial detection area into an electrical signal. The metasurface layer is a single-layer or double-layer structure. The sensitive functional layer is integrated with the metasurface layer. The metasurface layer modulates the light intensity at the preset spatial detection area by changing the optical properties of the optically tunable material, and focuses the incident light to the preset spatial detection area by reconstructing the phase distribution through the spatial light field. The geometric parameters and spatial orientation angle distribution of the composite unit structure are configured such that the optical response parameters of the system vary with the target environmental parameters, so as to establish a mapping relationship between the electrical signal and the target environmental parameters.

2. The integrated optical gas detection system based on a multifunctional metasurface according to claim 1, characterized in that, The metasurface layer is a single-layer structure, and the sensitive functional layer and the metasurface layer are composed of the same material layer. The composite unit structure encodes the spatial light field reconstructed phase distribution through its in-plane orientation angle distribution, and modulates the light intensity at the preset spatial detection area through the optical constant changes caused by the target environmental parameters.

3. The integrated optical gas detection system based on a multifunctional metasurface according to claim 1, characterized in that, The metasurface layer is a two-layer or multi-layer structure, which includes a sensitive functional layer, an isolation layer and a wavefront modulation layer in sequence along the incident light direction. The sensitive functional layer is used to respond to target environmental parameters and modulate the incident light field. The intermediate isolation layer is used to isolate and support the sensitive functional layer and the wavefront modulation layer. The wavefront modulation layer is used to form the spatial light field reconstruction phase distribution.

4. The integrated optical gas detection system based on a multifunctional metasurface according to claim 1, characterized in that, The optically tunable material includes at least one of palladium, palladium alloy, palladium-based multilayer composite film, transition metal oxide, or phase change material, and the target environmental parameter is any one of hydrogen concentration, methane concentration, or nitric oxide concentration.

5. The integrated optical gas detection system based on a multifunctional metasurface according to claim 1, characterized in that, The spatial light field reconstruction phase distribution includes an off-axis focusing phase component, which is configured to enhance the local light intensity of the incident light in the preset spatial detection area; the geometric parameters and / or in-plane orientation angle of the composite unit structure are calculated and determined according to the target phase distribution required by the preset spatial detection area.

6. The integrated optical gas detection system based on a multifunctional metasurface according to claim 1, characterized in that, The system operates in either reflective or transmissive modes, and the detected optical response parameters include any one of polarization conversion efficiency, transmittance, or reflectance.

7. The integrated optical gas detection system based on a multifunctional metasurface according to claim 1, characterized in that, The geometric parameters and spatial distribution of the composite unit structure are determined by an inverse optimization algorithm. The inverse optimization algorithm takes the spatial light field reconstruction phase distribution and the optical contrast of the sensitive functional layer under different environmental parameters as optimization targets, and performs optimization in a preset structural parameter space. The inverse optimization algorithm evaluates the resolvability and monotonicity of the response of different structures under environmental parameter changes through the quality factor, constrains the monotonicity of the optical response parameters under different environmental parameter states and the separation degree between adjacent states, so that the optical response parameters change with the target environmental parameters.

8. The integrated optical gas detection system based on a multifunctional metasurface according to claim 7, characterized in that, The reverse optimization algorithm includes an intelligent optimization strategy based on a surrogate model. The surrogate model includes a deep neural network, Gaussian process regression, or random forest. The global optimization mechanism of the intelligent optimization strategy includes at least one of Bayesian optimization framework, genetic algorithm, particle swarm optimization, topology optimization, gradient descent backpropagation, random search, or grid search.

9. The integrated optical gas detection system based on a multifunctional metasurface according to claim 7, characterized in that, The reverse optimization algorithm includes an adaptive switching mechanism, which dynamically adjusts the exploration and development weights of the acquisition function based on the uncertainty or iterative convergence state of the surrogate model during the optimization process, so as to balance global search and local refinement.

10. A control method for an integrated optical gas detection system based on a multifunctional metasurface, applied to the system according to any one of claims 1 to 9, characterized in that, Includes the following steps: After the incident light is modulated by the metasurface layer, the metasurface layer modulates the light intensity at the preset spatial detection area by changing the optical properties of the optically tunable material, and focuses the incident light to the preset spatial detection area by reconstructing the phase distribution through the spatial light field. The sensitive functional layer responds to changes in the target environmental parameters, and its optical properties undergo a reversible change, modulating the light intensity at the preset spatial detection area. The photodetector collects the optical signal at the preset spatial detection area and converts it into an electrical signal; The values ​​of the target environmental parameters are obtained by inversion based on the pre-calibrated correspondence between the electrical signal and the target environmental parameters.

Images