Arrayed waveguide grating spectrum chip based on two-dimensional function gradient and preparation method thereof

Abstract

The application discloses a two-dimensional function gradient-based array gradient filter spectrum chip and a preparation method thereof. The effective optical thickness of the filter structure continuously changes in X and Y directions, and the central transmission wavelength is detected. The detector array is in a row-column form. The wavelength difference between adjacent detection units in the same row is delta, the wavelength difference between corresponding columns of adjacent two rows is delta, one wavelength is mapped to one detection unit in a row-first manner, and the wavelengths of the detection units are different from each other. The application folds the single-axis arranged spectrum channel to a two-dimensional plane, and shortens the longest side of the chip in a single direction.

Description

Technical Field

[0001] This invention relates to the field of micro-spectral detection and thin-film optical device technology, and in particular to an array gradient filter spectral chip based on two-dimensional function gradient and its fabrication method. Background Technology

[0002] A linear graded-color filter is an optical thin-film device whose bandpass center wavelength changes continuously along a single direction. The center wavelength changes linearly along the X-axis while remaining constant along the Y-axis. By integrating linear graded-color filters with detector arrays, they form chip-level spectral detection systems, showing broad application prospects in portable spectral analysis, industrial online detection, medical diagnostics, and environmental monitoring.

[0003] In a single-axis linear graded filter, the spectral channels are arranged sequentially along a single graded direction. When a spectral chip provides multiple spectral channels, these channels are arranged in a row along the graded direction. The detector array requires the same number of detector units along the graded direction as the number of channels. The length of the chip along the graded direction is equal to the product of the number of channels and the spacing between the detector units. The spacing between the detector units is determined by the detector manufacturing process and has an incompressible physical lower limit. When the number of channels is large, the physical size of the single-axis linear graded filter along the graded direction increases linearly with the number of channels. The chip becomes too long along the graded direction, resulting in a long, thin strip shape and low area utilization. Existing technologies include schemes that splice or partition multiple single-axis linear graded filters, as well as discrete two-dimensional filter arrays, i.e., block mosaic structures. The center wavelengths of each filter block do not form a single, continuously varying mapping relationship along the two-dimensional direction. This fails to redistribute the spectral channel sequence arranged along a single direction to the detector units on the two-dimensional plane in a continuously varying manner, making it impossible to shorten the longest side of the chip along a single direction while maintaining a one-to-one correspondence between the spectral channels.

[0004] Therefore, this invention proposes an array gradient filter spectral chip based on two-dimensional function gradient and its fabrication method. Summary of the Invention

[0005] This invention provides a spectral chip based on a two-dimensional function gradient array of gradient filters and its fabrication method. By having a non-zero wavelength change rate in both the X and Y directions of the filter structure, the difference in the center transmission wavelength across two adjacent rows along the Y direction is N times the difference in the center transmission wavelength of adjacent detector units within the same row along the X direction. M×N center transmission wavelengths are mapped sequentially to M rows and N columns of detector units in a row-first manner, with each detector unit corresponding to a different center transmission wavelength. This invention folds the spectral channels of a uniaxial linear gradient filter, which need to be arranged along a single direction, into a two-dimensional detector unit array, shortening the longest side of the chip in a single direction from the spacing between M×N detector units to... With M· The larger one.

[0006] This invention provides a spectral chip based on a two-dimensional function gradient array of graded-pass filters, comprising a substrate, a two-dimensional continuously graded-pass filter structure, and a detector array optically coupled to the filter structure; the detector array is... OK A two-dimensional detector array, , All are positive integers greater than 1; The effective optical thickness of the filter structure varies continuously in both the X and Y directions, satisfying: ; The corresponding center transmission wavelength satisfies ; in, , Let the coordinates be those of the filter plane. For the initial thickness, , For thickness gradient coefficient, , These represent the wavelength change rates in the X and Y directions, respectively. and None of them are zero; The difference in center transmission wavelength between adjacent detector units within the same row along the X direction is ; The difference in center transmission wavelength between corresponding column detection units spanning two adjacent rows along the Y direction is: ; Each center transmission wavelength is mapped sequentially to in row priority. OK The array consists of detection units, each corresponding to a different center transmission wavelength.

[0007] Preferably, the filter structure covers a wavelength range from the first wavelength to the second wavelength. Arranged in order of priority along the row The center transmission wavelength is in constant step size Increasing, The spacing between the detection units in the X direction is The spacing of the detection units in the Y direction is Wavelength change rate in the X direction Wavelength change rate in the Y direction .

[0008] Preferably, the side length of the filter structure in the X direction is... The side length of the filter structure in the Y direction is The longest side of the chip in a single direction is and The larger of the two; a single-axis linear graded filter with the same wavelength range wavelength spacing with the same adjacent channel Cover the same number of The required gradient direction length for each spectral channel is The longest side is less than .

[0009] Preferably, the gradient function of the center transmission wavelength is: , The cross term coefficients cause the filter structure to generate a coupling thickness gradient in the X and Y directions, resulting in a non-linear distribution of the isowavelength lines of the center transmission wavelength.

[0010] Preferably, the filter structure is directly deposited on the detector array, or deposited on the substrate and then combined with the detector array; the detector array is any one of silicon detector array, indium gallium arsenide detector array, quantum dot detector array, germanium detector array, indium arsenide antimony detector array or thermoelectric detector array.

[0011] Preferably, the filter structure is a multilayer Fabry-Perot bandpass film system, with the thickness of each film layer varying in a continuous bidirectional gradient along the X and Y directions, and the film system is composed of alternating stacks of high-refractive-index film layers and low-refractive-index film layers.

[0012] Preferably, the filter structure is divided into multiple two-dimensional gradient regions on the substrate, and the gradient direction and gradient rate of each region are set independently. Each region maps its own spectral channel of its own band to its corresponding detector unit subarray in a row-first manner.

[0013] Preferably, the thickness gradient coefficient and The mask moving speed in the X direction is respectively related to the speed of the mask movement. and the mask movement speed in the Y direction satisfy , , For the film deposition rate, and Through independent adjustment and They are respectively set such that the difference in center transmission wavelength across two adjacent rows along the Y direction is equal to the difference in center transmission wavelength of adjacent detection units within the same row along the X direction. times.

[0014] This invention provides a method for fabricating an array gradient filter spectral chip based on a two-dimensional function gradient, comprising the following steps: Provide optical substrate; The substrate is placed above the deposition source in the vacuum coating chamber; During the deposition of the first refractive index film, the straight-edge mask is moved along the X direction at a velocity... The movement creates a thickness gradient in the first direction on the substrate, with a thickness gradient coefficient. With sedimentation rate and mask movement speed satisfy ; During the deposition of the second refractive index film, the straight-edge mask is made to move along the Y direction at a velocity The movement creates a second-direction thickness gradient on the substrate, with a thickness gradient coefficient. With sedimentation rate and mask movement speed satisfy ,adjust and This ensures that the difference in center transmission wavelengths between corresponding column detectors spanning two adjacent rows along the Y direction is equal to the difference in center transmission wavelengths between adjacent detectors within the same row along the X direction. times, This represents the number of detector array columns in the X direction; Alternating deposition of high-refractive-index and low-refractive-index films forms a multilayer Fabry-Perot bandpass filter structure. The bidirectional mask control steps described above are repeated when depositing each film layer, so that each film layer has a continuous thickness gradient in both the X and Y directions. The thickness gradients of each film layer are superimposed to form a two-dimensional continuous optical thickness distribution. Combine the filter structure with OK The detector array optical coupling of the array detector unit enables coverage of the target band as required. Each center transmission wavelength is mapped sequentially to in row priority. OK The detection unit of the column.

[0015] The advantages of this invention compared to existing technologies are as follows: it overcomes the technical shortcomings of existing single-axis linear graded filters, such as all spectral channels being arranged in a single direction, the chip's single-side size increasing linearly with the number of channels when the number of channels is large, and the chip being elongated and having low area utilization. This invention achieves this by making the filter structure have non-zero wavelength variation rates in both the X and Y directions, ensuring that the difference in center transmission wavelength across two adjacent rows along the Y direction is equal to the difference in center transmission wavelength of adjacent detector units within the same row along the X direction. Times, required to cover the target band Each center transmission wavelength is mapped sequentially to in row priority. OK The detector units are arranged in a row, each corresponding to a different central transmission wavelength, thus avoiding wavelength degeneracy. This invention utilizes detectors arranged in a single direction... Each spectral channel is folded into OK A two-dimensional detector array, the longest side of the chip in a single direction is formed by... The spacing between the detection units is shortened to and The larger one, while maintaining the same number of spectral channels and the same spectral resolution, transforms the spectral chip from a thin strip shape to a rectangle, thus improving the area utilization rate.

[0016] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0017] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of the row priority wavelength mapping process in an embodiment of the present invention; Figure 2 This is a schematic diagram of row-priority wavelength mapping in an embodiment of the present invention, in which M×N spectral channels arranged on a single axis are folded into a two-dimensional array of M rows and N columns; Figure 3 This is a schematic diagram of the overall structure of the array gradient filter spectral chip in an embodiment of the present invention; Figure 4 This is a schematic diagram of the row-priority center transmission wavelength distribution of the two-dimensional continuous gradient bandpass filter structure on the detector array in an embodiment of the present invention. (a) is the center transmission wavelength distribution of the M×N detector unit, (b) is a local magnification of the center transmission wavelength increasing along the X direction in a single row, and (c) is a continuous change curve of the center transmission wavelength increasing strictly linearly along the row priority number, with no jump at the row boundary, and the wavelengths of each detector unit being different. Detailed Implementation

[0019] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0020] like Figure 1 , Figure 2 , Figure 3 , Figure 4 As shown, this invention provides an embodiment of a two-dimensional function-gradient array gradient filter spectral chip, including a substrate, a two-dimensional continuously gradient bandpass filter structure, and a detector array optically coupled to the filter structure; the detector array is... OK A two-dimensional detector array, , All are positive integers greater than 1; The effective optical thickness of the filter structure varies continuously in both the X and Y directions, satisfying: ; The corresponding center transmission wavelength satisfies ; in, , Let the coordinates be those of the filter plane. For the initial thickness, , For thickness gradient coefficient, , These represent the wavelength change rates in the X and Y directions, respectively. and None of them are zero; The difference in center transmission wavelength between adjacent detector units within the same row along the X direction is ; The difference in center transmission wavelength between corresponding column detection units spanning two adjacent rows along the Y direction is: ; Each center transmission wavelength is mapped sequentially to in row priority. OK The array consists of detection units, each corresponding to a different center transmission wavelength.

[0021] In this embodiment, the two-dimensional continuously graded bandpass filter structure is a bandpass filter with continuously varying wavelengths in both the X and Y directions. The transmission wavelength at the bandpass center changes continuously with the planar position coordinates. This structure is formed by depositing optical thin films layer by layer on a substrate. Effective optical thickness. This refers to the position of each film layer in the filter structure. The cumulative optical thickness at a given point is equal to the sum of the products of the geometric thicknesses of each film layer and their refractive indices. The central transmission wavelength is proportional to the effective optical thickness; therefore, the central transmission wavelength exhibits the following characteristics: The filter structure is a two-dimensional linear distribution. The substrate is made of a transparent material and supports the filter structure. The detector array is a two-dimensional array device formed by arranging multiple independent photodetector units in rows and columns. Each detector unit corresponds to a local position on the filter structure, used to receive a specific wavelength of light signal transmitted through the filter structure at that position and convert it into an electrical signal output. The wavelength change rate A characterizes the change in the central transmitted wavelength per unit distance along the X-direction, in nanometers per millimeter. The wavelength change rate B characterizes the change in the central transmitted wavelength per unit distance along the Y-direction, in nanometers per millimeter. Both A and B being non-zero means that the filter structure exhibits wavelength gradients in both the X and Y directions.

[0022] In this embodiment, the detector array is assumed to be OK The spacing between the detection units along the X direction is... The spacing between the detection units along the Y direction is , and Equal or unequal. Located at the th Line number The center coordinates of the detection unit of the column are , where i is 0 to integers, From 0 to The integer. The center transmission wavelength corresponding to this detection unit is... Let the difference in center transmission wavelength between adjacent detector units along the X-direction within the same row be denoted as . ,Right now To ensure that the center transmission wavelengths required to cover the target band are arranged continuously and without repetition along the row priority order, the difference in center transmission wavelengths corresponding to the detector cells spanning two adjacent rows along the Y direction is: ,Right now Therefore, the center transmission wavelength corresponding to the detection unit in the i-th row and j-th column is... ,in For this detection unit in OK The one-dimensional indexes in the column array are arranged in row-major order, with values ​​ranging continuously from 0 to... .

[0023] The physical meaning of row priority mapping is: within the same row, adjacent detector cells along the X direction have their center transmission wavelengths successively differing by a certain amount. When starting from the last column of a certain row (the... When a column moves to the first column (column 0) of the next row, the row number... Add 1 to the column number Depend on When the wavelength becomes 0, the change in the center transmission wavelength is This is exactly equal to the wavelength interval between adjacent detector elements within the same row. Therefore, the entire OK The center transmission wavelength on the column array increases continuously with a constant step size δ along the row priority order. Each detection unit corresponds to a serial number from 0 to... of Each detector unit has a unique central transmission wavelength, and there is no wavelength degeneracy. The step size δ is a design value determined based on the required band span and number of detector units for the actual application, and can be any positive value.

[0024] In this embodiment, if a single-axis linear graded filter is to provide the same... There are 10 spectral channels, and these channels must be arranged in a row along the gradient direction. The corresponding detector array needs to be aligned in that direction. There are one detection unit, and the chip has a length of [missing information] in this direction. The present invention folds these channels into OK A two-dimensional array of columns, with the chip having a side length of 0 in the X direction. The side length in the Y direction is The longest side of the chip in a single direction is and The larger one is much smaller than the single-axis scheme. While maintaining the same number of spectral channels and the same wavelength spacing between adjacent channels, this invention transforms the spectral chip from a long, thin strip shape into a rectangle, increasing the spectral channel density per unit area. The chip is square when Since the chip is rectangular, both fall within the protection scope of this invention.

[0025] In a specific design example, the detector array is arranged in 12 rows and 20 columns, with a detector element spacing of 25 micrometers along both the X and Y directions, and a wavelength spacing δ between adjacent channels designed to be 4 nanometers. The total number of spectral channels is then 240, with a wavelength range of... Nanometers. Wavelength change rate in the X direction. nanometers per millimeter, wavelength change rate in the Y direction Nanometers per millimeter. The side length of the chip in the X direction is... Millimeters, with a side length of in the Y direction. The longest side is 0.50 mm. A single-axis linear graded filter requires a graded direction length of [missing information - likely a unit of measurement] millimeters to provide the same 240 channels. Millimeters. This invention shortens the longest side of the chip from 6.00 mm to 0.50 mm. The above values ​​are design parameters used in the theoretical calculations of the technical solution according to this invention, and are intended to clearly illustrate the technical principles and beneficial effects of this invention through specific numerical examples, and do not constitute a limitation on the scope of protection of this invention.

[0026] This embodiment is a spectral chip for the visible to near-infrared band. A glass substrate is used. High-refractive-index films are made of tantalum pentoxide or titanium dioxide, while low-refractive-index films are made of silicon dioxide. Tantalum pentoxide and silicon dioxide, or titanium dioxide and silicon dioxide, are alternately deposited on the glass substrate to form a multilayer Fabry-Perot bandpass film system. The thickness of each film layer exhibits a continuous bidirectional gradient along both the X and Y directions. An indium gallium arsenide (IGaAs) detector array is used, covering a response band from 900 nm to 1700 nm. The center transmission wavelength of the filter structure is designed to cover the response band of the IGaAs detector array, from 900 nm to 1700 nm, with a band span of 800 nm. The glass substrate and the tantalum pentoxide / silicon dioxide or titanium dioxide / silicon dioxide film system exhibit good transmittance and sufficient refractive index difference within this band, forming narrow-band transmission characteristics, which, combined with the IGaAs detector array, enables spectral detection in the near-infrared band.

[0027] In this embodiment, the filter structure is fabricated using vacuum deposition. A deposition source and a straight-edge mask are placed in the vacuum deposition chamber, with the substrate positioned above the deposition source. During the deposition of the first refractive index film, the straight-edge mask is moved along the X-direction at a velocity... The material moves, creating a thickness gradient in the first direction on the substrate. During the deposition of the second refractive index film, the straight-edge mask is moved along the Y direction at a velocity... The process involves moving the film to create a second-direction thickness gradient on the substrate. A multilayer Fabry-Perot bandpass filter structure is formed by alternately depositing high-refractive-index and low-refractive-index layers. The bidirectional masking control steps described above are repeated during the deposition of each layer, ensuring a continuous gradient distribution of geometric thickness in both the X and Y directions for each layer. The superposition of these geometric thickness gradients results in a two-dimensional continuous distribution of the effective optical thickness.

[0028] In this embodiment, the first refractive index film is either a high refractive index film or a low refractive index film, and the second refractive index film is either a high refractive index film or a low refractive index film. The refractive index of the high refractive index film is greater than that of the low refractive index film. The material of the high refractive index film can be tantalum pentoxide or titanium dioxide, and the material of the low refractive index film can be silicon dioxide. A straight-edge mask is a shielding component placed between the deposition source and the substrate during the deposition process. This shielding component has a straight edge. By changing the projection position of this straight edge on the substrate surface, different locations on the substrate are exposed to the deposition source for different times, thereby forming a continuous gradient distribution of film thickness on the substrate. The straight-edge mask is mounted on a two-dimensional translation stage, which includes an X-direction drive mechanism and a Y-direction drive mechanism. Each drive mechanism independently receives motion control commands, allowing the mask to move independently along the X and Y directions during the deposition process, thereby forming thickness gradients in the X and Y directions respectively during the deposition of each film layer. Mask moving speed. and These refer to the moving speeds of the straight-edge blocking mask along the X and Y directions, respectively. Thickness gradient coefficient. Proportional to deposition rate With mask moving speed The ratio of the thickness gradient coefficient Proportional to deposition rate With mask moving speed The ratio of . and respectively with and To satisfy a direct proportional relationship, the mask moving speed can be adjusted independently. and To set separately and Then set them separately and Deposition rate This refers to the thickness of the film deposited on the substrate surface per unit time, a parameter determined by the coating process conditions. The movement trajectory of the straight-edge mask along the X and Y directions can both be straight lines, or at least one can be an arc to create a non-linear thickness gradient distribution, thus forming a layer containing cross terms. The central transmission wavelength distribution, where the cross-term coefficients... The degree of coupling between the thickness gradient distributions in the X and Y directions is determined by the degree of coupling, which is adjusted by the radius of curvature of the arc of the mask movement trajectory. A nonlinear thickness gradient distribution refers to a film thickness change along a certain direction that is not constant but varies with position. This distribution is formed by using an arc-trajectory mask to make the rate of change of the exposure time of different positions of the substrate to the deposition source non-constant. This embodiment, through the above-described preparation method, achieves a bidirectional continuous gradient change in the thickness of each film layer along both the X and Y directions, and the superposition of the thickness gradients of each film layer forms a two-dimensional continuous optical thickness distribution, with the effective optical thickness continuously varying in both the X and Y directions. The geometric thickness gradient superposition of each film layer refers to the superposition of the thickness gradients of each film layer at the same position... The effective optical thickness at a given location is obtained by summing the geometric thicknesses of each layer, multiplying the refractive index of each layer by its geometric thickness, and then summing the results. In this embodiment, the Fabry-Perot bandpass filter structure consists of alternately stacked high-refractive-index and low-refractive-index layers. The thickness of each layer varies continuously along both the X and Y directions, ensuring that the central transmission wavelength of the entire filter structure varies continuously in both directions. By controlling the thickness distribution of each layer, the central transmission wavelength is made to satisfy… .

[0029] In this embodiment, in the scheme of forming multiple two-dimensional gradient regions on the substrate by partitioning, a light-shielding layer can be set in the spacer area between each region to prevent optical crosstalk between adjacent regions. The gradient direction and gradient rate of each region are realized by setting the mask moving direction and moving speed corresponding to each region respectively.

[0030] In this embodiment, the first and second wavelengths are selected according to specific application requirements. When applied to visible-near-infrared spectroscopy detection, the first wavelength can be 400 nm and the second wavelength can be 900 nm. When applied to near-infrared spectroscopy detection, the first wavelength can be 900 nm and the second wavelength can be 1700 nm. In specific applications, the first and second wavelengths are selected according to actual spectral detection requirements. For example, when applied to visible-near-infrared spectroscopy detection, the first wavelength can be 400 nm and the second wavelength can be 900 nm; when applied to near-infrared spectroscopy detection, the first wavelength can be 900 nm and the second wavelength can be 1700 nm.

[0031] In this embodiment, the number of detector units in the detector array and the positional resolution on the filter structure jointly determine the spectral resolution of the spectral chip. In this invention, the thickness gradients in the X and Y directions, the Fabry-Perot film system structure, and the bidirectional mask-moving fabrication method have a close synergistic relationship. Specifically, the Fabry-Perot film system forms narrow-band transmission characteristics through the alternating stacking of multiple high-refractive-index and low-refractive-index layers, providing the basic structure for spectral selection of two-dimensional continuous gradients. The bidirectional mask-moving method independently controls the mask movement speed in the X and Y directions during the deposition of each film layer, ensuring that each film layer in the Fabry-Perot film system continuously varies along both the X and Y directions, thereby accurately transferring the two-dimensional gradient distribution to each film layer. The superposition of the thickness gradients of each film layer forms a two-dimensional continuous distribution of the total effective optical thickness, allowing M×N detector units to correspond to different central transmission wavelengths. These three technical features are interdependent and work together; the absence of any one of them would prevent the achievement of the technical effect of "folding spectral channels arranged in a single direction into a two-dimensional array and shortening the longest side of the chip in a single direction" in this invention. If there is only a two-dimensional thickness gradient distribution without a Fabry-Perot film system, a bandpass filter structure with wavelength selection function cannot be formed; if there is only a Fabry-Perot film system with infinite bidirectional movement control of the mask, a continuous thickness gradient distribution in two dimensions cannot be formed; if there are only the former two without the superposition and synergy of the thickness gradients of each film layer, the two-dimensional distribution of effective optical thickness cannot be accurately realized at each film layer level.

[0032] like Figure 3 As shown in the figure, this is a cross-sectional structural diagram of the array gradient filter spectral chip in an embodiment of the present invention. The figure shows the stacking relationship of the array gradient filter structure, detector array, readout circuit layer and shell.

[0033] like Figure 4 As shown, this is a schematic diagram of the row-priority center transmission wavelength distribution of the two-dimensional continuous gradient bandpass filter structure on the detector array in an embodiment of the present invention, wherein (a) is The distribution of the center transmission wavelength of the detection unit, (b) is a local magnification of the center transmission wavelength increasing along the X direction in a single row, and (c) is a curve showing the continuous change of the center transmission wavelength along the row priority number.

[0034] In another embodiment of the invention, the side length of the filter structure in the X direction is... The side length of the filter structure in the Y direction is The longest side of the chip in a single direction is and The larger of the two; a single-axis linear graded filter with the same wavelength range wavelength spacing with the same adjacent channel Cover the same number of The required gradient direction length for each spectral channel is The longest side is less than

[0035] In another embodiment of the invention, the gradient function of the center transmission wavelength is: , The cross term coefficients cause the filter structure to generate a coupling thickness gradient in the X and Y directions, resulting in a non-linear distribution of the isowavelength lines of the center transmission wavelength.

[0036] In this embodiment, the gradient function of the central transmission wavelength refers to a mathematical function describing the relationship between the central transmission wavelength at various positions on the filter structure and the X and Y coordinates. This function is used to determine the value of the central transmission wavelength at any position of the filter structure in the plane. The wavelength change rate in the X direction represents the amount of change of the central transmission wavelength per unit distance along the X direction, and this parameter independently controls the wavelength gradient rate in the X direction. The wavelength change rate in the Y direction represents the amount of change of the central transmission wavelength per unit distance along the Y direction, and this parameter independently controls the wavelength gradient rate in the Y direction. The cross term coefficient represents the rate of change of the central transmission wavelength with the product of the X and Y coordinates. This coefficient makes the wavelength changes in the X and Y directions no longer independent, but rather produces a coupling effect.

[0037] In this embodiment, the coupled thickness gradient refers to a nonlinear thickness distribution formed by the superposition of thickness gradients in both the X and Y directions when thickness changes occur simultaneously. In a linear gradient scheme with only a first-order term (i.e., the wavelength change rate in the X direction multiplied by the X coordinate and the wavelength change rate in the Y direction multiplied by the Y coordinate), the wavelength change rate in the X and Y directions is constant, and the thickness gradients in the two directions are independent of each other. The slope of the thickness change along the Y direction is the same at any X coordinate, and the slope of the thickness change along the X direction is the same at any Y coordinate. The thickness gradient is uniform everywhere in the plane. When a cross term is introduced (i.e., the cross term coefficient is multiplied by the X coordinate and then by the Y coordinate), the thickness gradient of the filter structure is no longer uniform. The magnitude and direction of the thickness gradient change continuously with the plane position. This change is manifested as the thickness gradient in the X direction being modulated by the Y coordinate, and the thickness gradient in the Y direction being modulated by the X coordinate, thus forming a coupling between the thickness gradients in the two directions.

[0038] In this embodiment, the formation mechanism of the coupled thickness gradient lies in the fact that when the thickness distribution includes cross terms, the rate of change of thickness with respect to the X direction is equal to the rate of change of wavelength in the X direction plus the cross term coefficient multiplied by the Y coordinate. This rate of change varies linearly with the Y coordinate, indicating that the thickness gradient along the X direction has different values ​​at different Y positions. Similarly, the rate of change of thickness with respect to the Y direction is equal to the rate of change of wavelength in the Y direction plus the cross term coefficient multiplied by the X coordinate. This rate of change also varies linearly with the X coordinate, indicating that the thickness gradient along the Y direction has different values ​​at different X positions. This phenomenon of thickness gradient varying with position results in the filter structure having different wavelength gradient rates and gradient directions in different regions, and the isowavelength lines of the central transmission wavelength form a curved distribution in the plane according to the value of the cross term coefficient.

[0039] In this embodiment, the cross-term coefficient can be positive, negative, or zero. When the cross-term coefficient is zero, this embodiment degenerates into a linear two-dimensional gradient scheme. When the cross-term coefficient is positive, the wavelength change rate in the X direction increases along the positive Y direction, and the wavelength change rate in the Y direction increases along the positive X direction, resulting in a divergent distribution of the thickness gradient in the plane. When the cross-term coefficient is negative, the wavelength change rate in the X direction decreases along the positive Y direction, and the wavelength change rate in the Y direction decreases along the positive X direction, resulting in a convergent distribution of the thickness gradient in the plane.

[0040] In this embodiment, the absolute value of the cross-term coefficient determines the strength of the coupling effect. The larger the absolute value of the cross-term coefficient, the more significant the modulation of the thickness gradient in the X direction by the Y coordinate, and the more significant the modulation of the thickness gradient in the Y direction by the X coordinate, resulting in a wavelength distribution in the plane that deviates more from a linear distribution. By adjusting the value of the cross-term coefficient, the wavelength spatial distribution of the spectral chip can be customized for specific application requirements. For example, the chip can achieve higher wavelength resolution in the central region and a wider band coverage in the edge region, or the wavelength can be spread along a specific curved path to match the non-rectangular arrangement of the detector array.

[0041] In this embodiment, the gradient function containing cross terms and the linear two-dimensional gradient function belong to different technical solutions. The linear two-dimensional gradient function has a constant wavelength change rate in both the X and Y directions, and the thickness gradient is uniform and consistent in direction throughout the plane. In this embodiment, the gradient function containing cross terms no longer has a constant wavelength change rate in the X and Y directions, but changes continuously with the plane position. The magnitude and direction of the thickness gradient change throughout the plane, thereby forming a nonlinear gradient filter structure.

[0042] In this embodiment, the gradient function containing cross terms alters the shape of the isowavelength lines through coupled thickness gradients in the X and Y directions. In the linear two-dimensional gradient scheme, the isowavelength lines are straight lines. In the nonlinear gradient scheme of this embodiment, the isowavelength lines are curves, and the shape of these curves is determined by the values ​​of the cross term coefficients. When the cross term coefficients are positive, the isowavelength lines bend in the positive Y direction, and the degree of bending increases with the absolute value of the cross term coefficients. When the cross term coefficients are negative, the isowavelength lines bend in the negative Y direction.

[0043] In this embodiment, the gradient function containing cross terms can be achieved by making at least one of the movement trajectories of the straight-edge mask along the X and Y directions an arc trajectory during the fabrication process. An arc trajectory refers to the curved shape of the projection path of the mask edge onto the substrate surface; this curve can be upward-convex, downward-concave, or S-shaped. When the mask moves along the arc trajectory, the rate of change of time of exposure of different positions on the substrate to the deposition source is no longer a constant value, but changes continuously with the mask position. This causes the change in the thickness of the film deposited on the substrate along the movement direction to vary with position, forming a nonlinear thickness gradient distribution. The bending direction and degree of the arc trajectory determine the sign and absolute value of the cross term coefficients. When the mask moves linearly along the X direction and arc-shaped along the Y direction, the thickness gradient in the X direction remains constant, while the thickness gradient in the Y direction varies with the X coordinate. The cross term coefficients are formed by the contribution of the arc trajectory in the Y direction. When the mask moves arc-shaped along both the X and Y directions, the thickness gradients in both directions vary with position, forming a higher-order nonlinear gradient distribution.

[0044] This embodiment introduces a cross-term coefficient to generate a coupling thickness gradient between the X and Y directions in the filter structure, thereby expanding the wavelength spatial distribution from linear to nonlinear. This makes the wavelength arrangement of the spectral chip more flexible and adaptable to different spectral detection needs and application scenarios.

[0045] There are two ways to combine the filter structure with the detector array. First, the filter structure is directly deposited onto the detector array. This involves forming a two-dimensional, continuously graded bandpass filter structure in situ on the surface of the detector array. The bottom layer of the filter structure is in direct contact with the photosensitive surface of the detector array, without any additional transparent substrate or spacer layer. This eliminates the extra thickness space occupied by the substrate at the bottom of the filter structure and avoids alignment errors and light energy loss introduced during optical coupling after separate fabrication. Second, the filter structure is deposited on a separate substrate to form a filter. This filter is then combined with the detector array through alignment bonding, optical adhesive bonding, or mechanical fixation. This ensures that each local position on the filter structure corresponds one-to-one with each detection unit on the detector array. This method facilitates separate fabrication and testing of the filter structure and detector before assembly and is compatible with existing detector devices.

[0046] In this embodiment, the detector array is a variety of detector arrays with a response band in the ultraviolet to mid-infrared range, including but not limited to silicon detector arrays, indium gallium arsenide detector arrays, quantum dot detector arrays, germanium detector arrays, indium arsenide antimony detector arrays, and thermoelectric detector arrays. Silicon detector arrays are suitable for the ultraviolet to near-infrared band, with a response band typically covering 200 nm to 1100 nm. Indium gallium arsenide detector arrays are suitable for the near-infrared band, with a response band typically covering 900 nm to 1700 nm, and extended indium gallium arsenide detectors can extend to 2600 nm. Germanium detector arrays are suitable for the visible to near-infrared band. Quantum dot detector arrays can be adapted to different bands from the visible to the infrared by adjusting the quantum dot size. Indium arsenide antimony detector arrays are suitable for the mid-infrared band, with a response band typically covering 3 μm to 5 μm. Thermoelectric detector arrays are suitable for a wide infrared band and do not require cooling. By selecting different types of detector arrays and configuring a two-dimensional continuously gradient bandpass filter structure that matches their response band, the spectral chip of the present invention can cover different spectral bands from the ultraviolet to the mid-infrared.

[0047] This embodiment is a mid-infrared spectral chip. The substrate is a calcium fluoride substrate, which is transparent over a wide wavelength range of 0.2 μm to 8 μm, making it suitable for the mid-infrared band. High-refractive-index and low-refractive-index films suitable for the mid-infrared band are deposited on the calcium fluoride substrate. The high-refractive-index film is made of germanium, and the low-refractive-index film is made of yttrium fluoride, barium fluoride, or zinc sulfide. Germanium and yttrium fluoride, or germanium and zinc sulfide, are deposited alternately to form a multilayer Fabry-Perot bandpass film system, with the thickness of each film exhibiting a continuous bidirectional gradient along both the X and Y directions. The detector array is an indium arsenide (IAM) detector array, covering a response band of 3 μm to 5 μm. The center transmission wavelength of the filter structure is designed to cover the response band of the IAM detector array, ranging from 3000 nm to 5000 nm, with a band span of 2000 nm. Calcium fluoride substrates and germanium / fluoride or germanium / zinc sulfide films exhibit good light transmittance and a large refractive index difference in the mid-infrared band, forming narrow-band transmission characteristics. When combined with an indium arsenide antimony detector array, they can achieve spectral detection in the mid-infrared band.

[0048] In another embodiment of the present invention, the filter structure is a multilayer Fabry-Perot bandpass film system, wherein the thickness of each film layer varies in a bidirectional continuous gradient along the X and Y directions, and the film system is composed of alternating stacks of high refractive index film layers and low refractive index film layers.

[0049] In this embodiment, the Fabry-Perot bandpass film system refers to an optical thin film structure system composed of two reflective film layers and a spacer layer between the two reflective film layers. This structure system utilizes the principle of multi-beam interference to allow light signals of a specific wavelength to pass through while reflecting light signals of other wavelengths. When incident light enters the Fabry-Perot bandpass film system, the light signal undergoes multiple reflections and transmissions between the two reflective film layers. The wavelength of light signal that satisfies the coherence enhancement condition obtains the maximum transmittance, and this wavelength is the center transmission wavelength at that location. The wavelength of light signal that does not satisfy the coherence enhancement condition is reflected or consumed.

[0050] In this embodiment, the central transmission wavelength of the Fabry-Perot bandpass film system is equal to twice the effective optical thickness of the spacer layer divided by the interference order. At the basic interference order, the effective optical thickness of the spacer layer is equal to half the central transmission wavelength. When the effective optical thickness of the spacer layer changes continuously along the X and Y directions, the central transmission wavelength changes synchronously and continuously along the X and Y directions.

[0051] In this embodiment, the bidirectional continuous gradient variation of film thickness along the X and Y directions means that the geometric thickness of each layer in the Fabry-Perot bandpass film system, including the high-refractive-index and low-refractive-index layers in the reflective layers and the spacer layer, exhibits a continuous gradient distribution in both the X and Y directions, without any abrupt changes or step-like jumps in thickness. The two-dimensional continuous distribution pattern.

[0052] In this embodiment, a high-refractive-index film layer refers to a film layer made of a high-refractive-index material, with a relatively high refractive index value. A low-refractive-index film layer refers to a film layer made of a low-refractive-index material, with a relatively low refractive index value. The refractive index of the high-refractive-index film layer is greater than that of the low-refractive-index film layer. The refractive index difference between the high-refractive-index and low-refractive-index film layers forms an optical interface, which reflects the incident light signal. Multiple high-refractive-index and low-refractive-index film layers are alternately stacked to form a periodic reflective film layer structure, which has high reflectivity within a specific wavelength range.

[0053] In this embodiment, alternating stacking refers to the sequential deposition of high-refractive-index and low-refractive-index films in the order of high-refractive-index, low-refractive-index, high-refractive-index, and low-refractive-index. Adjacent layers are made of different materials, with alternating high and low refractive indices, forming a multilayer optical interference film system. The thickness of each layer in the system continuously varies along the X and Y directions, causing the reflection and transmission spectra of the entire system to continuously change with the planar position.

[0054] In this embodiment, the Fabry-Perot bandpass film system comprises at least two reflective layers and one spacer layer. Each reflective layer is composed of multiple sets of alternating high-refractive-index and low-refractive-index layers. The spacer layer is disposed between the two reflective layers, and its effective optical thickness determines the position of the central transmission wavelength. The number of alternating sets of high-refractive-index and low-refractive-index layers in the reflective layer determines the reflectivity of the reflective layer; the more sets, the higher the reflectivity, the narrower the transmission peak, and the higher the spectral resolution. The reflectivity of the reflective layer increases with the number of layers and approaches a limiting value.

[0055] In this embodiment, the thickness of each film layer exhibits a continuous bidirectional gradient along both the X and Y directions. This is achieved by controlling the movement of the straight-edge mask along both the X and Y directions during the deposition of each film layer. During the deposition of a high-refractive-index film layer, the straight-edge mask first moves along the X direction at a first velocity to form a thickness gradient distribution in the X direction, and then moves along the Y direction at a second velocity to form a thickness gradient distribution in the Y direction. Similarly, during the deposition of a low-refractive-index film layer, the straight-edge mask similarly first moves along the X direction at a third velocity to form a thickness gradient distribution in the X direction, and then moves along the Y direction at a fourth velocity to form a thickness gradient distribution in the Y direction. The thickness gradient distributions of the high-refractive-index and low-refractive-index films in the X and Y directions are superimposed, resulting in an effective optical thickness. Thickness gradient coefficient and It is contributed by the thickness gradient of each film layer.

[0056] In this embodiment, the high-refractive-index film is made of tantalum pentoxide or titanium dioxide, both of which have higher refractive indices than the low-refractive-index film in the visible to near-infrared bands. The low-refractive-index film is made of silicon dioxide. The refractive index difference between the high-refractive-index and low-refractive-index films forms an optical interface that reflects the incident light signal. The greater the refractive index difference, the higher the reflectivity and the sharper the transmission peak of the reflective film formed by alternating stacking with the same number of layers.

[0057] In this embodiment, the continuous variation of the thickness of each film layer along the X and Y directions is achieved by controlling the moving speed of the straight-edge mask along the X and Y directions. The thickness gradient coefficient is proportional to the ratio of the deposition rate to the mask moving speed. Each film layer can have the same or different thickness gradient coefficients. When the thickness gradient coefficients of each film layer are the same, the proportion of thickness change with position is consistent, and the change in effective optical thickness is entirely contributed by the synchronous change of the thickness of each film layer. When the thickness gradient coefficients of each film layer are different, the proportion of thickness change with position is inconsistent, and the relationship between effective optical thickness and single-layer film thickness varies with position, enabling more complex wavelength distribution patterns.

[0058] This embodiment forms a two-dimensional continuous gradient bandpass filter structure with narrow-band transmission characteristics by alternately stacking high-refractive-index and low-refractive-index layers of a multilayer Fabry-Perot bandpass film system with a bidirectional continuous gradient change in the X and Y directions. This ensures the monochromaticity of the transmission spectrum at each position and achieves a two-dimensional continuous distribution of the central transmission wavelength.

[0059] In another embodiment of the present invention, the filter structure is divided into multiple two-dimensional gradient regions on the substrate, and the gradient direction and gradient rate of each region are set independently. Each region maps the spectral channels of its respective band to its corresponding detector unit subarray in a row-first manner.

[0060] In this embodiment, partitioning to form multiple two-dimensional gradient regions refers to dividing the same substrate into multiple independent regions. Each region is equipped with a filter structure that has a two-dimensional continuous gradient bandpass filtering function. The regions are separated from each other by spacers or light-shielding layers, and different regions have different central transmission wavelength distribution ranges. Each region is equivalent to an independent spectral chip unit, and multiple spectral chip units are integrated on the same substrate.

[0061] In this embodiment, the independent setting of the gradient direction and gradient rate for each region means that the filter structure of each region has its own independent wavelength change rate in the X and Y directions. The wavelength change rate values ​​of different regions can be the same or different. The X and Y wavelength change rates of the first region are set independently, and the X and Y wavelength change rates of the second region are also set independently. The wavelength change rates of the first and second regions do not affect each other. Each region maps its respective spectral channel of its own band to its corresponding detector unit subarray in a row-first manner.

[0062] In this embodiment, the regions are separated from each other by interval regions. These interval regions do not have filter structures, and their width is determined based on actual process capabilities and optical crosstalk suppression requirements. A light-shielding layer, made of opaque material, can be provided in the interval regions to prevent light signals from transmitting from one region into the corresponding detector unit of an adjacent region, thus avoiding optical crosstalk between adjacent regions.

[0063] In this embodiment, the band ranges covered by each region can be the same or different. The first region covers the first band span from the first wavelength to the second wavelength, and the second region covers the second band span from the third wavelength to the fourth wavelength. The first and second band spans can partially overlap, not overlap at all, or be completely identical. When each region covers different band ranges, multiple discrete bands can be detected simultaneously on a single spectral chip, which is suitable for applications that require simultaneous detection of multiple discontinuous spectral ranges. When each region covers the same band range, multiple parallel detection channels can be formed on a single spectral chip, improving the reliability of the detection results through redundant detection of the same band range by multiple regions.

[0064] In this embodiment, the dimensions of each region can be the same or different. The side lengths of the first region in the X and Y directions are determined by the region's band span and a set wavelength variation rate, and the side lengths of the second region in the X and Y directions are also determined by the region's band span and a set wavelength variation rate. The dimensions of each region are set independently and do not affect each other.

[0065] In this embodiment, the gradient direction and gradient rate of each region are independently set by setting independent straight-edge masking masks above each region. During the deposition of the first film layer, the straight-edge masking mask corresponding to the first region moves along the X and Y directions at a first speed group, and the straight-edge masking mask corresponding to the second region moves along the X and Y directions at a second speed group. The straight-edge masking masks of each region are independently controlled in terms of movement direction and speed during the deposition process, thereby forming different thickness gradient distributions in each region. The partitioned masking technology is achieved by forming multiple independent mask opening regions on the same mask plate. Each mask opening region corresponds to a gradient region, and the masking edges of each mask opening region can move independently.

[0066] In this embodiment, the detector array corresponding to each region can be a group of local detection units corresponding to that region within a whole detector array, or it can be multiple independent detector array chips, each corresponding to a region. When a whole detector array is used, the filter structure of each region is directly deposited on different regions of the same detector array, and the detector array regions corresponding to the gaps between regions do not receive optical signals or are shielded after receiving optical signals.

[0067] This embodiment forms multiple two-dimensional gradient regions on the substrate, enabling a single spectral chip to simultaneously cover multiple spectral bands or realize multiple parallel detection channels, thus expanding the application scenarios and detection capabilities of the spectral chip. At the same time, each region maps its respective spectral channels to its corresponding detection unit subarray in a row-first manner, so that the size of each region is smaller than the size required to cover the same spectral band under the single-axis gradient scheme.

[0068] In another embodiment of the invention, the thickness gradient coefficient and The mask moving speed in the X direction is respectively related to the speed of the mask movement. and the mask movement speed in the Y direction satisfy , , For the film deposition rate, and Through independent adjustment and They are respectively set such that the difference in center transmission wavelength across two adjacent rows along the Y direction is equal to the difference in center transmission wavelength of adjacent detection units within the same row along the X direction. times.

[0069] In this embodiment, the thickness gradient coefficient refers to the rate of change of the effective optical thickness along the X direction in a two-dimensional continuously graded bandpass filter structure. It represents the amount of change in the effective optical thickness per unit distance in the X direction, and this coefficient determines the rate of change of the central transmission wavelength along the X direction. The thickness gradient coefficient also refers to the rate of change of the effective optical thickness along the Y direction in a two-dimensional continuously graded bandpass filter structure. It represents the amount of change of the effective optical thickness per unit distance in the Y direction, and this coefficient determines the rate of change of the central transmission wavelength along the Y direction.

[0070] In this embodiment, the deposition rate refers to the film thickness deposited on the substrate surface per unit time during vacuum deposition. This parameter is determined by the evaporation rate or sputtering rate of the deposition source in the deposition equipment. Under the same deposition process conditions, the deposition rate remains constant. The higher the deposition rate, the greater the film thickness deposited per unit time.

[0071] In this embodiment, the X-direction mask moving speed refers to the speed at which the straight-edge blocking mask moves along the X-direction. This speed determines the time difference in exposure of different locations on the substrate to the deposition source along the X-direction. When the mask moves slowly along the X-direction, the time difference in exposure to the deposition source at different locations on the substrate along the X-direction is large, resulting in a large thickness gradient. When the mask moves quickly along the X-direction, the time difference in exposure to the deposition source at different locations on the substrate along the X-direction is small, resulting in a small thickness gradient.

[0072] In this embodiment, the mask moving speed in the Y direction refers to the speed at which the straight-edge blocking mask moves along the Y direction. This speed determines the time difference in exposure of different locations on the substrate to the deposition source along the Y direction. When the mask moves slowly along the Y direction, the time difference in exposure to the deposition source at different locations on the substrate along the Y direction is large, resulting in a large thickness gradient. When the mask moves quickly along the Y direction, the time difference in exposure to the deposition source at different locations on the substrate along the Y direction is small, resulting in a small thickness gradient.

[0073] In this embodiment, the thickness gradient coefficient is proportional to the ratio of the deposition rate to the mask moving speed in the X direction. Specifically, the thickness gradient coefficient is equal to a proportionality constant multiplied by the deposition rate and then divided by the mask moving speed in the X direction. This proportionality constant is determined by the geometry of the deposition equipment and the shading effect at the mask edges, and remains a constant value once the equipment is fixed. When the deposition rate increases, the film thickness deposited per unit time at the same mask moving speed increases, and the thickness gradient coefficient increases accordingly. When the mask moving speed in the X direction increases, the time required for the mask to sweep across a unit distance on the substrate surface decreases, the thickness difference along the X direction decreases, and the thickness gradient coefficient decreases accordingly.

[0074] In this embodiment, the thickness gradient coefficient is proportional to the ratio of the deposition rate to the mask moving speed in the Y direction. Specifically, the thickness gradient coefficient is equal to a proportionality constant multiplied by the deposition rate and then divided by the mask moving speed in the Y direction. This proportionality constant is determined by the geometry of the deposition equipment and the shading effect at the mask edges, and remains a constant value once the equipment is fixed. When the deposition rate increases, the film thickness deposited per unit time at the same mask moving speed increases, and the thickness gradient coefficient increases accordingly. When the mask moving speed in the Y direction increases, the time required for the mask to sweep across a unit distance on the substrate surface decreases, the thickness difference along the Y direction decreases, and the thickness gradient coefficient decreases accordingly.

[0075] In this embodiment, the thickness gradient coefficient is set independently by adjusting the mask moving speed in the X and Y directions. This means that the thickness gradient coefficient in the X direction is determined solely by the mask moving speed in the X direction combined with the deposition rate, and is independent of the mask moving speed in the Y direction. Similarly, the thickness gradient coefficient in the Y direction is determined solely by the mask moving speed in the Y direction combined with the deposition rate, and is independent of the mask moving speed in the X direction. Therefore, by changing the mask moving speed in the X direction, the thickness gradient coefficient in the X direction can be adjusted independently without affecting the thickness gradient coefficient in the Y direction. Similarly, by changing the mask moving speed in the Y direction, the thickness gradient coefficient in the Y direction can be adjusted independently without affecting the thickness gradient coefficient in the X direction. This independent control of the thickness gradient coefficient allows the wavelength change rates in the X and Y directions to be set independently, without being mutually constrained.

[0076] In this embodiment, by independently adjusting the mask moving speed in the X direction and the mask moving speed in the Y direction, the thickness gradient coefficients in the X and Y directions can be set respectively, and then the wavelength change rates A and B in the X and Y directions can be set respectively, so that the difference in the center transmission wavelength across two adjacent rows along the Y direction is N times the difference in the center transmission wavelength of adjacent detection units in the same row along the X direction.

[0077] In this embodiment, the thickness gradient coefficient is directly proportional to the wavelength change rate in the X direction. This proportionality is determined by the interference conditions of the Fabry-Perot bandpass filter structure, and the proportionality coefficient is equal to twice the equivalent refractive index. Similarly, the thickness gradient coefficient is directly proportional to the wavelength change rate in the Y direction, also determined by the interference conditions of the Fabry-Perot bandpass filter structure, with the proportionality coefficient equal to twice the equivalent refractive index. Therefore, adjusting the thickness gradient coefficient is equivalent to adjusting the wavelength change rate in the X direction, and adjusting the thickness gradient coefficient is equivalent to adjusting the wavelength change rate in the Y direction.

[0078] In this embodiment, the independent adjustment of the mask moving speed in the X and Y directions is achieved by setting the motion parameters of the X-direction drive mechanism and the Y-direction drive mechanism respectively. The X-direction drive mechanism controls the moving speed of the straight-edge masking mask along the X direction, and the Y-direction drive mechanism controls the moving speed of the straight-edge masking mask along the Y direction. The two drive mechanisms receive motion control commands independently and do not interfere with each other. The adjustment range of the mask moving speed is determined according to the motion control accuracy of the coating equipment, and can typically be continuously adjusted within the range of 0.1 mm per minute to 100 mm per minute.

[0079] This embodiment achieves independent quantitative setting of the thickness gradient coefficients in the X and Y directions by independently adjusting the mask moving speeds in the X and Y directions, thereby independently setting the wavelength change rate in the X and Y directions, making the wavelength distribution parameters of the two-dimensional gradient filter structure designable and controllable.

[0080] This invention provides an embodiment of a method for fabricating an array gradient filter spectral chip based on a two-dimensional function gradient, comprising the following steps: Provide optical substrate; The substrate is placed above the deposition source in the vacuum coating chamber; During the deposition of the first refractive index film, the straight-edge mask is moved along the X direction at a velocity... The movement creates a thickness gradient in the first direction on the substrate, with a thickness gradient coefficient. With sedimentation rate and mask movement speed satisfy ; During the deposition of the second refractive index film, the straight-edge mask is made to move along the Y direction at a velocity The movement creates a second-direction thickness gradient on the substrate, with a thickness gradient coefficient. With sedimentation rate and mask movement speed satisfy ,adjust and This ensures that the difference in center transmission wavelengths between corresponding column detectors spanning two adjacent rows along the Y direction is equal to the difference in center transmission wavelengths between adjacent detectors within the same row along the X direction. times, This represents the number of detector array columns in the X direction; Alternating deposition of high-refractive-index and low-refractive-index films forms a multi-layer Fabry-Perot bandpass filter structure. The bidirectional mask control steps described above are repeated when depositing each film layer, so that each film layer has a continuous thickness gradient in both the X and Y directions. The thickness gradients of each film layer are superimposed to form a two-dimensional continuous optical thickness distribution. Combine the filter structure with OK The detector array optical coupling of the array detector unit enables coverage of the target band as required. Each center transmission wavelength is mapped sequentially to in row priority. OK The detection unit of the column.

[0081] In this embodiment, the optical substrate refers to a light-transmitting substrate used to support the filter structure. The substrate is made of a light-transmitting material, such as quartz glass, silicon wafer, sapphire, or other materials that are transparent within the target wavelength range. The upper surface of the substrate is optically polished to provide a smooth surface for the deposition of the filter structure.

[0082] The substrate is placed above the deposition source in the vacuum coating chamber.

[0083] In this embodiment, the vacuum coating chamber refers to a sealed container used for the vacuum coating process. The interior of the chamber can be evacuated to a predetermined vacuum level, typically less than one-thousandth of a Pascal. A deposition source is located below the interior of the chamber, and the deposition source contains coating material for forming the film. The deposition source evaporates or sputters the coating material by heating or electron beam bombardment, forming a stream of material particles flying towards the substrate. The substrate is located above the deposition source, with its lower surface facing the deposition source. The material particles evaporated or sputtered from the deposition source are deposited on the lower surface of the substrate to form the film.

[0084] During the deposition of the first refractive index film, the straight-edge mask is moved along the X direction at a speed to form a first-direction thickness gradient on the substrate. The thickness gradient coefficient is proportional to the deposition rate and the mask moving speed. The thickness gradient coefficient is equal to the proportionality constant multiplied by the deposition rate and divided by the mask moving speed in the X direction.

[0085] In this embodiment, the first refractive index film is either a high refractive index film or a low refractive index film. A straight-edge shielding mask is disposed between the deposition source and the substrate, having a straight edge extending along the Y direction and partially shielding the substrate surface along the X direction. Material particles evaporated or sputtered from the deposition source are partially blocked by the straight-edge shielding mask, while some reach the substrate surface for deposition through the exposed area at the mask edge. As the straight-edge shielding mask moves at a speed along the X direction, the exposed area on the substrate changes over time, and the substrate surface at different X positions is exposed to the deposition source for different durations, thereby forming a continuous gradient distribution of film thickness along the X direction on the substrate. The speed in the X direction controls the rate of change of film thickness along the X direction; the slower the speed, the greater the rate of change, and the faster the speed, the smaller the rate of change.

[0086] During the deposition of the second refractive index film, the straight-edge mask is moved along the Y direction at a speed to form a second-direction thickness gradient on the substrate. The thickness gradient coefficient is proportional to the deposition rate and the mask moving speed. The thickness gradient coefficient is equal to the proportionality constant multiplied by the deposition rate and divided by the mask moving speed in the Y direction.

[0087] In this embodiment, the second refractive index film is either a high refractive index film or a low refractive index film. When the first refractive index film is a high refractive index film, the second refractive index film is a low refractive index film. When the first refractive index film is a low refractive index film, the second refractive index film is a high refractive index film. The straight edge of the straight-edge masking mask extends along the X direction during the deposition of the second refractive index film, partially blocking the substrate surface along the Y direction. When the straight-edge masking mask moves at a speed along the Y direction, a continuous gradient distribution of film thickness is formed on the substrate along the Y direction. The speed in the Y direction controls the rate of change of film thickness along the Y direction; the slower the speed, the greater the rate of change, and the faster the speed, the smaller the rate of change.

[0088] Alternating deposition of high-refractive-index and low-refractive-index films forms a multilayer Fabry-Perot bandpass filter structure. The bidirectional mask control steps described above are repeated when depositing each film layer, so that each film layer has a continuous thickness gradient in both the X and Y directions. The thickness gradients of each film layer are superimposed to form a two-dimensional continuous optical thickness distribution.

[0089] In this embodiment, alternating deposition refers to the sequential deposition of multiple films in the order of high-refractive-index film, low-refractive-index film, high-refractive-index film, and low-refractive-index film, with adjacent films made of different materials and alternating refractive indices. Each film layer independently undergoes the aforementioned bidirectional mask control steps during deposition. Specifically, during the deposition of each high-refractive-index film layer, a straight-edge mask is moved along the X-direction at an X-direction velocity to form an X-direction thickness gradient, and then moved along the Y-direction at a Y-direction velocity to form a Y-direction thickness gradient. Similarly, during the deposition of each low-refractive-index film layer, a straight-edge mask is moved along the X-direction at an X-direction velocity to form an X-direction thickness gradient, and then moved along the Y-direction at a Y-direction velocity to form a Y-direction thickness gradient. The superposition of the X-direction thickness gradients of each film layer results in a continuous variation of the total effective optical thickness along the X-direction, and the superposition of the Y-direction thickness gradients of each film layer results in a continuous variation of the total effective optical thickness along the Y-direction. The total effective optical thickness formed by the superposition of the thickness gradients of each film layer satisfies a two-dimensional continuous distribution, wherein the thickness gradient coefficient in the X direction is determined by the sum of the thickness gradient contributions of each film layer in the X direction, and the thickness gradient coefficient in the Y direction is determined by the sum of the thickness gradient contributions of each film layer in the Y direction.

[0090] In this embodiment, optical coupling refers to assembling the filter structure with the detector array so that the detector array can effectively receive the specific wavelength light signal transmitted by the filter structure. Optical coupling methods include directly depositing the filter structure onto the photosensitive surface of the detector array, or fixing the filter structure formed on a separate substrate to the photosensitive surface of the detector array using optical adhesive or mechanical alignment. After optical coupling, each local position on the filter structure forms a one-to-one spatial relationship with each detection unit on the detector array, and each detection unit receives the specific wavelength light signal transmitted by the filter structure at its corresponding position.

[0091] This fabrication method controls the bidirectional movement of the straight-edge mask along the X and Y directions, independently forming thickness gradients in the X and Y directions during the deposition of each film layer. By superimposing the thickness gradients of each film layer, a two-dimensional continuous distribution of effective optical thickness is formed, realizing the controllable fabrication of array gradient filter spectral chips based on two-dimensional function gradients.

[0092] During the deposition of the first refractive index film, a straight-edge mask is moved at a speed along the X direction to form a first-direction thickness gradient on the substrate. This step, by superimposing with the subsequent Y-direction thickness gradient, provides an X-direction gradient component for achieving a two-dimensional continuous thickness gradient distribution. The X-direction thickness gradient formed alone cannot constitute the two-dimensional gradient structure of this invention.

[0093] During the deposition of the second refractive index film, a straight-edge mask is moved along the Y direction at a certain speed, forming a second-direction thickness gradient on the substrate. This step, together with the aforementioned X-direction thickness gradient, constitutes the basis of a biaxial gradient in orthogonal directions. The thickness gradient coefficients α and β in the two directions are independently set by controlling the mask moving speeds v1 and v2, respectively, so that the wavelength change rates A and B in the X and Y directions can be independently adjusted.

[0094] Alternating deposition of high-refractive-index and low-refractive-index films forms a multilayer Fabry-Perot bandpass filter structure. The bidirectional mask control steps described above are repeated during the deposition of each film layer. In this step, the bidirectional thickness gradient of each film layer is formed independently, and the superposition of the thickness gradients of each film layer contributes to the two-dimensional continuous distribution of the total effective optical thickness.

[0095] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A spectral chip based on a two-dimensional function gradient array of gradient filters, characterized in that, It includes a substrate, a two-dimensional continuously graded bandpass filter structure, and a detector array optically coupled to the filter structure; the detector array is... OK A two-dimensional detector array, , All are positive integers greater than 1; The effective optical thickness of the filter structure varies continuously in both the X and Y directions, satisfying: ； The corresponding center transmission wavelength satisfies ; in, , Let the coordinates be those of the filter plane. For the initial thickness, , For thickness gradient coefficient, , These represent the wavelength change rates in the X and Y directions, respectively. and Neither of them is zero; The difference in center transmission wavelength between adjacent detector units within the same row along the X direction is ; The difference in center transmission wavelength between corresponding column detection units spanning two adjacent rows along the Y direction is: ; Each center transmission wavelength is mapped sequentially to in row priority. OK The array consists of detection units, each corresponding to a different center transmission wavelength.

2. The array gradient filter spectral chip based on two-dimensional function gradient as described in claim 1, characterized in that, The filter structure covers a wavelength range from the first wavelength to the second wavelength. Arranged in order of priority along the row The center transmission wavelength is in constant step size Increasing, ; The spacing between the detection units in the X direction is The spacing of the detection units in the Y direction is Wavelength change rate in the X direction Wavelength change rate in the Y direction .

3. The array gradient filter spectral chip based on two-dimensional function gradient as described in claim 2, characterized in that, The side length of the filter structure in the X direction is The side length of the filter structure in the Y direction is The longest side of the chip in a single direction is and The larger of the two; a single-axis linear graded filter with the same wavelength range wavelength spacing with the same adjacent channel Cover the same number of The required gradient direction length for each spectral channel is The longest side is less than .

4. The array gradient filter spectral chip based on two-dimensional function gradient as described in claim 1, characterized in that, The gradient function of the center transmission wavelength is , The cross term coefficient causes the filter structure to generate a coupling thickness gradient in the X and Y directions, and the isowavelength lines of the center transmission wavelength are distributed non-linearly.

5. The array gradient filter spectral chip based on two-dimensional function gradient as described in claim 1, characterized in that, The filter structure is deposited directly on the detector array, or deposited on the substrate and then combined with the detector array; the detector array is any one of silicon detector array, indium gallium arsenide detector array, quantum dot detector array, germanium detector array, indium arsenide antimony detector array or thermoelectric detector array.

6. The array gradient filter spectral chip based on two-dimensional function gradient according to claim 1, characterized in that, The filter structure is a multilayer Fabry-Perot bandpass film system, with the thickness of each film layer varying continuously in both directions along the X and Y directions. The film system is composed of alternating stacks of high-refractive-index and low-refractive-index film layers.

7. The array gradient filter spectral chip based on two-dimensional function gradient according to claim 1, characterized in that, The filter structure is divided into multiple two-dimensional gradient regions on the substrate. The gradient direction and gradient rate of each region are set independently. Each region maps its own spectral channel to its corresponding detector unit subarray in a row-first manner.

8. The array gradient filter spectral chip based on two-dimensional function gradient according to claim 1, characterized in that, Thickness gradient coefficient and The mask moving speed in the X direction is respectively related to the speed of the mask movement. and the mask movement speed in the Y direction satisfy , , For the film deposition rate, and Through independent adjustment and They are respectively set such that the difference in center transmission wavelength across two adjacent rows along the Y direction is equal to the difference in center transmission wavelength of adjacent detection units within the same row along the X direction. times.

9. A method for fabricating a spectral chip based on a two-dimensional function gradient array of gradient filters, characterized in that, Includes the following steps: Provide optical substrate; The substrate is placed above the deposition source in the vacuum coating chamber; During the deposition of the first refractive index film, the straight-edge mask is moved along the X direction at a velocity... The movement creates a thickness gradient in the first direction on the substrate, with a thickness gradient coefficient. With sedimentation rate and mask movement speed satisfy ; During the deposition of the second refractive index film, the straight-edge mask is made to move along the Y direction at a velocity The movement creates a second-direction thickness gradient on the substrate, with a thickness gradient coefficient. With sedimentation rate and mask movement speed satisfy ,adjust and This ensures that the difference in center transmission wavelengths between corresponding column detectors spanning two adjacent rows along the Y direction is equal to the difference in center transmission wavelengths between adjacent detectors within the same row along the X direction. times, This represents the number of detector array columns in the X direction; Alternating deposition of high-refractive-index and low-refractive-index films forms a multi-layer Fabry-Perot bandpass filter structure. The bidirectional mask control steps described above are repeated when depositing each film layer, so that each film layer has a continuous thickness gradient in both the X and Y directions. The thickness gradients of each film layer are superimposed to form a two-dimensional continuous optical thickness distribution. Combine the filter structure with OK The detector array optical coupling of the array detector unit enables coverage of the target band as required. Each center transmission wavelength is mapped sequentially to in row priority. OK The detection unit of the column.

Images