Talbot self-imaging-based optical waveguide master mask defect self-adaptive photoetching method

Abstract

The invention relates to the technical field of photoetching, in particular to an optical waveguide master mask defect self-adaptive photoetching method based on Talbot self-imaging, which comprises the following steps of: calculating a Talbot self-imaging distance by collecting a design period and an exposure wavelength parameter of an optical waveguide master mask, and driving a substrate coated with photoresist to be accurately positioned to the distance; after positioning is completed, coherent light output is started, meanwhile, the substrate is controlled to perform in-plane periodic motion, distortion components caused by defects are optically counteracted by a diffraction light field in the propagation process through the Fresnel diffraction effect, and homogenized light intensity distribution is generated; and after the distribution is recorded by the photoresist, a grating structure is formed through development, and the optical waveguide template is prepared through etching and deposition processes. Defect suppression can be realized on the optical level, the structural integrity and preparation consistency of the template are improved, the scrap demand of the mother set is reduced, and the production cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of photolithography, and in particular to an adaptive photolithography method for defects in optical waveguide master plates based on Tyber self-imaging. Background Technology

[0002] In the application of optical waveguide template photolithography, Tabo self-imaging is an optical phenomenon that uses periodic diffraction to generate repeating self-imaging patterns, providing a high-resolution exposure template for the photolithography process. The optical waveguide master is the original template used to replicate the optical waveguide structure, and its fabrication accuracy is related to the performance of the optical waveguide device. Defect-adaptive photolithography is a photolithography method that compensates for manufacturing defects by monitoring and adjusting exposure parameters in real time. Combined with the periodic characteristics of Tabo self-imaging, pattern generation can be optimized during the fabrication of the optical waveguide master, reducing the impact of defects and thus improving the fabrication quality of the template.

[0003] Existing adaptive photolithography technology for optical waveguide master defects has the following technical pain points: In existing near-contact or contact photolithography processes, micron or nanoscale defects on the surface of the optical waveguide master, such as scratches, broken lines, or periodic disorders, are directly transferred to the photoresist template. The defect transfer process lacks an effective suppression mechanism. For example, scratches on the master can lead to the formation of corresponding grooves on the template surface, and incomplete patterns on the master can cause missing grating lines on the template. Such defect transfer causes damage to the integrity of the template structure, and optical waveguide devices fabricated based on defective templates suffer from performance degradation problems such as increased waveguide loss or decreased diffraction efficiency. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an adaptive photolithography method for optical waveguide master template defects based on Taibo self-imaging. This invention solves the technical problem of optical waveguide performance degradation caused by the inability to effectively suppress the propagation of defects in the replication template.

[0005] To solve the above-mentioned technical problems, the specific contents of the present invention are as follows:

[0006] This invention provides an adaptive photolithography method for defects in optical waveguide master plates based on Tiber self-imaging, comprising: Step 1: Collect the design cycle value of the optical waveguide master and the coherent light wavelength value for exposure, and generate a parameter data package; Step 2: Extract the design cycle value and wavelength value from the parameter data package, and perform calculations on the design cycle value and wavelength value to obtain the Tiber self-imaging distance value. Step 3: Convert the Talbot self-imaging distance value into a displacement control signal to drive the substrate coated with photoresist to move. When the difference between the measured distance between the substrate and the master plate and the Talbot self-imaging distance value is less than the set threshold, the positioning is determined to be complete. Step 4: Once the positioning is complete, start the coherent light output and simultaneously control the substrate to perform periodic in-plane motion with preset frequency and amplitude parameters. Step 5: Coherent light passes through the optical waveguide mother plate to form a diffracted light field. During the propagation of the diffracted light field to the substrate position, the distortion component caused by the defect is canceled by the Fresnel diffraction effect. The distortion component is the non-uniform part of the diffracted light field caused by the defect. Step 6: The photoresist records the light intensity distribution after the cancellation process, and the photoresist is treated with a developer to form a grating structure. Then, the optical waveguide template is prepared by etching and deposition of waveguide functional layers.

[0007] Furthermore, in the adaptive photolithography method for optical waveguide master plate defects based on Tiber self-imaging described in this invention, step 1, which involves acquiring the design period value and the coherent light wavelength value used for exposure of the optical waveguide master plate and generating a parameter data package, includes: Read the period markings in the optical waveguide master design drawing to generate the first design period value, or use an optical measurement device to scan the surface of the master, collect the original period signal, perform digital filtering on the original period signal, and output the second design period value. The nominal wavelength value of the laser source is collected, and the actual spectrum of the laser output is monitored using a spectrometer. The nominal wavelength value is compared with the center wavelength of the actual spectrum, and the wavelength value is output. The first or second design period value is encapsulated with the wavelength value to generate a parameter data package.

[0008] Furthermore, in the adaptive lithography method for optical waveguide master defects based on Tiger self-imaging described in this invention, step 2, which involves extracting design period values ​​and wavelength values ​​from the parameter data package and performing calculations on the design period values ​​and wavelength values ​​to obtain Tiger self-imaging distance values, includes: Receive parameter data packets and extract the design period value and wavelength value from the parameter data packets; The separated design period values ​​are squared to generate the period square value, and the separated wavelength values ​​are reciprocal to generate the wavelength reciprocal value. Multiplying the squared value of the period by the reciprocal of the wavelength yields a preliminary calculation result for the Talber distance; Using the preliminary calculation result of the Talbot distance as the initial distance parameter, optical diffraction simulation calculation is performed. If the light field uniformity value output by the simulation calculation is lower than the preset standard value, the initial distance parameter is iteratively adjusted until the light field uniformity value reaches the preset standard value, and the final Talbot self-imaging distance value is output.

[0009] Furthermore, in the adaptive photolithography method for waveguide master defects based on Talbot self-imaging described in this invention, step 3, in which the Talbot self-imaging distance value is converted into a displacement control signal to drive the substrate coated with photoresist to move, and the positioning is determined to be complete when the difference between the measured distance between the substrate and the master and the Talbot self-imaging distance value is less than a set threshold, includes: Read the self-imaging distance value of the Titan and generate a displacement control signal based on the self-imaging distance value of the Titan; The displacement control signal controls the substrate to move towards the optical waveguide mother plate; During the movement, the real-time distance between the base surface and the mother plate surface is measured to obtain the real-time distance measurement value; Calculate the absolute difference between the real-time distance measurement value and the Taibo self-imaging distance value, and determine whether the absolute difference is less than a set threshold. If the absolute difference is greater than or equal to the set threshold, a feedback control signal is generated based on the sign of the difference. The feedback control signal is superimposed on the displacement control signal, so that the value of the displacement control signal is adjusted according to the difference ratio, guiding the base to continue moving. If the absolute difference is less than the set threshold, the feedback control signal will stop being generated and a positioning completion status signal will be output.

[0010] Furthermore, in the adaptive photolithography method for optical waveguide master defects based on Talbot self-imaging described in this invention, step 4, in the state of completed positioning, involves activating the coherent light source output and simultaneously controlling the substrate to perform periodic in-plane motion with a preset frequency and amplitude, including: Receive the positioning completion status signal; In response to the positioning completion status signal, coherent light is generated and modulated to form a uniform illumination field; Based on the photosensitive properties of the photoresist and the period of the target grating structure, the substrate motion frequency parameters and amplitude parameters are determined; The base is controlled to perform in-plane reciprocating motion according to frequency and amplitude parameters; During the reciprocating motion, the actual displacement data of the base is monitored, and the actual displacement data is compared with the frequency parameters and amplitude parameters. If the deviation exceeds the tolerance, the amplitude parameter in the motion control parameters is adjusted proportionally according to the magnitude of the deviation.

[0011] Furthermore, in the adaptive photolithography method for optical waveguide master plate defects based on Talbot self-imaging described in this invention, in step 5, the coherent light passes through the optical waveguide master plate to form a diffracted light field. When the diffracted light field propagates to the substrate position, it cancels the distortion component caused by the defect through the Fresnel diffraction effect, including: The uniform illumination field formed in step 4 is received, and the uniform illumination field illuminates the defective optical waveguide mother plate. The periodic structure and defects on the optical waveguide mother plate jointly modulate the illumination light field, generating a diffracted light field, which includes information about the mother plate structure and defect information. The diffracted light field propagates from the optical waveguide mother plate to the substrate. Multiple acquisition points are set along the propagation path of the diffracted light field, and the phase data and amplitude data of the diffracted light field at the acquisition points are acquired by a photodetector. Perform Fourier transform on the phase data and amplitude data to convert the diffracted light field into a spatial frequency spectrum, and identify the high-frequency components corresponding to the periodic structure of the master plate and the low-frequency components corresponding to the defect distortion from the spatial frequency spectrum. Based on the principle of Taybernating self-imaging, high-frequency components coherently superimpose at the Taybernating self-imaging distance due to periodic matching, increasing the light intensity to more than 1.5 times the original value. Low-frequency components diverge at the Taybernating self-imaging distance due to non-periodic characteristics, decreasing the light intensity to less than 30% of the original value. The enhanced high-frequency components and the attenuated low-frequency components are combined at the substrate location to obtain a uniform light intensity distribution.

[0012] Furthermore, in the adaptive photolithography method for optical waveguide master defects based on Tyber self-imaging described in this invention, step 6, in which the photoresist records the light intensity distribution after cancellation processing, and forms a grating structure after treatment with a developing solution, includes: The uniform light intensity distribution obtained in step 5 is propagated to the substrate surface coated with photoresist through an optical system; The uniform light intensity distribution acts on the photoresist layer, causing the photosensitive components in the photoresist to absorb the photon energy in the light intensity distribution; After absorbing photon energy, the photosensitive component undergoes a photochemical reaction, which leads to cross-linking or decomposition of the photosensitive component in the exposed area; Based on the spatial intensity variation of light intensity distribution, a grating pattern is formed in the photoresist layer that is potential on the surface. In the region where the light intensity is higher than the photoresist photosensitive threshold, a complete photochemical reaction occurs, while in the region where the light intensity is lower than the photoresist photosensitive threshold, an incomplete photochemical reaction occurs. The developer is applied to the photoresist surface to dissolve the areas of photoresist that have not undergone a complete photochemical reaction, while preserving the areas of photoresist that have undergone a complete photochemical reaction. After dissolution, a grating structure corresponding to the uniform light intensity distribution is formed on the photoresist surface.

[0013] Furthermore, in the adaptive photolithography method for optical waveguide master defects based on Talbot self-imaging described in this invention, step 6, the further etching and deposition of waveguide functional layers to complete the fabrication of the optical waveguide template, includes: The surface morphology of the grating structure formed after development was measured to obtain grating groove depth data, line width data, and period size data. Based on the grating groove depth data, line width data, and period size data, calculate the etching time parameters, etching gas flow rate parameters, and power parameters; Etching is performed according to etching time parameters, etching gas flow rate parameters, and power parameters to transfer the grating pattern on the photoresist surface into the substrate material; Waveguide functional materials are deposited on the substrate surface after pattern transfer, according to preset thickness and refractive index requirements; The optical performance of the template after deposition of waveguide functional material is tested, and the transmission loss and diffraction efficiency are measured. The measurement results are compared with the performance of the template prepared by the defect-free master, and the defect suppression effect evaluation data are output.

[0014] Furthermore, the adaptive photolithography method for optical waveguide master plate defects based on Talbot self-imaging described in this invention also includes: In step 3, the current distance data between the base and the master plate is obtained in real time; in step 4, the base motion frequency data and amplitude data are obtained in real time. Establish the correspondence between distance data and motion frequency data and amplitude data; When the difference between the current distance data and the Tiber self-imaging distance value is less than twice the set threshold, the motion amplitude data will be reduced to 50% of its original value. When the base motion trajectory deviates, adjust the feedback control signal in step 3; The generation of the positioning completion status signal simultaneously satisfies both the distance threshold condition and the motion stability condition.

[0015] Furthermore, the adaptive photolithography method for optical waveguide master plate defects based on Talbot self-imaging described in this invention also includes: The value of the self-imaging distance of the Titanium laser output in step 2 is used as the basis for generating the displacement control signal in step 3. The positioning completion status signal generated in step 3 serves as the trigger condition for starting exposure in step 4. The substrate motion start time in step 4 is synchronized with the coherent light output start time; Monitor the self-imaging distance value, actual distance measurement value, motion frequency data, motion amplitude data, and light intensity distribution data of the Titanium laser. If any data exceeds the preset normal range threshold, the process is paused and a specific adjustment instruction is generated. The adjustment instruction includes the data name, deviation amount, and adjustment direction. The process continues after the data returns to the normal range threshold.

[0016] Beneficial effects of this invention: The present invention provides an adaptive photolithography method for optical waveguide master defects based on Talbot self-imaging. By utilizing the self-imaging effect in the Fresnel diffraction process, when the photoresist substrate is located at the Talbot distance, the distortion component caused by defects is averaged by the strong signal of the surrounding intact period during the propagation of the diffracted light field, thereby achieving defect suppression at the optical level. This allows a uniform light intensity distribution to be formed on the photoresist even if there are micron or nanometer-scale defects in the master, ultimately obtaining a structurally complete optical waveguide template. This avoids the template performance degradation caused by direct defect propagation in existing processes, reduces the need for master scrapping and re-fabrication, lowers production costs, and improves the reliability of the photolithography process and the consistency of template quality. Attached Figure Description

[0017] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on the drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of an adaptive photolithography method for optical waveguide master plate defects based on Taibo self-imaging according to the present invention. Detailed Implementation

[0019] To make the technical solution of the present invention clearer, the present invention will be clearly and completely described below with reference to specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. The present invention provided by various embodiments will be described in detail below with reference to the accompanying drawings. To better understand the purpose of the present invention, the present invention will be described in further detail below.

[0020] Please see Figure 1 The present invention provides an adaptive photolithography method for defects in optical waveguide master plates based on Talbot self-imaging, comprising: Step 1: Collect the design cycle value of the optical waveguide master and the coherent light wavelength value for exposure, and generate a parameter data package; Step 2: Extract the design cycle value and wavelength value from the parameter data package, and perform calculations on the design cycle value and wavelength value to obtain the Tiber self-imaging distance value. Step 3: Convert the Talbot self-imaging distance value into a displacement control signal to drive the substrate coated with photoresist to move. When the difference between the measured distance between the substrate and the master plate and the Talbot self-imaging distance value is less than the set threshold, the positioning is determined to be complete. Step 4: Once the positioning is complete, start the coherent light output and simultaneously control the substrate to perform periodic in-plane motion with preset frequency and amplitude parameters. Step 5: Coherent light passes through the optical waveguide mother plate to form a diffracted light field. During the propagation of the diffracted light field to the substrate position, the distortion component caused by the defect is canceled by the Fresnel diffraction effect. The distortion component is the non-uniform part of the diffracted light field caused by the defect. Step 6: The photoresist records the light intensity distribution after the cancellation process, and the photoresist is treated with a developer to form a grating structure. Then, the optical waveguide template is prepared by etching and deposition of waveguide functional layers.

[0021] The adaptive lithography method for optical waveguide master plates based on Tiber self-imaging achieves defect suppression through a systematic process. In the initial stage, the design period value and the wavelength value of the coherent light used for exposure of the optical waveguide master plate are acquired to generate a parameter data package. The design period value can be read from the master plate design drawing or obtained by scanning the master plate surface with optical measurement equipment to acquire the original period signal, which is then output after digital filtering. The wavelength value is determined by comparing the nominal wavelength value of the laser source with the actual spectral center wavelength monitored by the spectrometer. The encapsulated parameter data package provides the basic data support for subsequent calculations.

[0022] After the design period and wavelength values ​​are extracted from the parameter data package, calculations are performed on these values ​​to obtain the Tiger self-imaging distance. The calculation process includes squaring the design period value, reciprocally calculating the wavelength value, and multiplying the results to generate a preliminary distance value. For further optimization, optical diffraction simulation calculations can be performed, using the preliminary distance value as the initial parameter for iterative adjustment until the light field uniformity reaches a preset standard, at which point the final Tiger self-imaging distance value is output.

[0023] The self-imaging distance value from the TIGAP sensor is converted into a displacement control signal, which drives the photoresist-coated substrate to move. The displacement control signal guides the substrate towards the waveguide master plate, while simultaneously measuring the distance between the substrate and the master plate in real time. When the absolute difference between the real-time distance measurement value and the TIGAP self-imaging distance value is less than a set threshold, the system determines that positioning is complete and outputs a positioning completion status signal.

[0024] Once positioning is complete, coherent light output is activated, and the beam is modulated to form a uniform illumination field. The substrate is synchronously controlled to perform periodic in-plane motion at a preset frequency and amplitude. The motion parameters are set based on the photoresist's photosensitive characteristics and the target grating structure's periodicity. During motion, the actual displacement data of the substrate is monitored and compared with the preset frequency and amplitude parameters. If the deviation exceeds the tolerance, the amplitude parameters are adjusted proportionally to maintain motion stability.

[0025] Coherent light passes through the optical waveguide master plate, forming a diffracted light field. During its propagation to the substrate, it utilizes the Fresnel diffraction effect. The diffracted light field includes information about the periodic structure of the master plate and distortion components caused by defects. By acquiring phase and amplitude data at multiple spatial locations, a Fourier transform is performed to convert it into a spatial frequency spectrum. At the Talbot self-imaging distance, high-frequency components are enhanced due to periodically matched coherent superposition, while low-frequency components are attenuated due to aperiodic divergence, thus canceling out distortion components and generating a uniform light intensity distribution.

[0026] Photoresist records a uniform light intensity distribution. Photosensitive components absorb energy and undergo photochemical reactions, resulting in cross-linking or decomposition of exposed areas to form potential grating patterns. The developer dissolves insufficiently reacted areas while retaining the fully reacted areas, forming a complete grating structure. Subsequent etching processes transfer the grating pattern to the substrate material, and the waveguide functional layer is deposited to complete the optical waveguide template fabrication. Template optical performance is tested by measuring transmission loss and diffraction efficiency, and the defect suppression effect is evaluated by comparing it with the results of defect-free master template fabrication.

[0027] Specifically, in the adaptive photolithography method for optical waveguide master plate defects based on Tiber self-imaging described in this invention, step 1, which involves acquiring the design period value and the coherent light wavelength value used for exposure of the optical waveguide master plate and generating a parameter data package, includes: Read the period markings in the optical waveguide master design drawing to generate the first design period value, or use an optical measurement device to scan the surface of the master, collect the original period signal, perform digital filtering on the original period signal, and output the second design period value. The nominal wavelength value of the laser source is collected, and the actual spectrum of the laser output is monitored using a spectrometer. The nominal wavelength value is compared with the center wavelength of the actual spectrum, and the wavelength value is output. The first or second design period value is encapsulated with the wavelength value to generate a parameter data package.

[0028] The process of generating the parameter data package involves acquiring the design period values ​​of the optical waveguide master and the coherent light wavelength values ​​used for exposure. This step provides fundamental data support for subsequent calculations of the Tiber self-imaging distance, and the reliability of the input parameters is ensured through a standardized process.

[0029] One method to obtain design period values ​​is by reading the period markings in the waveguide master design drawings. Design drawings typically include period parameters for the grating structure. The first design period value is generated by extracting the period markings through file analysis or manual identification. This method relies on the accuracy of the drawings and is suitable when the parameters are known during the master design stage. Another method uses optical measurement equipment to scan the master surface and acquire raw period signals. Optical measurement equipment, such as high-resolution microscopes or diffractometers, performs non-contact scanning of the master to obtain raw data on the surface structure. The raw period signal is then digitally filtered to remove noise and interference, outputting a second design period value. The digital filtering algorithm uses low-pass or band-pass filtering to enhance the signal-to-noise ratio and ensure the accuracy of the period measurement. These two methods complement each other; the design drawings provide theoretical values, while optical measurements provide actual values, adapting to different application scenarios.

[0030] The process involves acquiring the nominal wavelength of the laser source while simultaneously monitoring the actual spectrum of the laser output using a spectrometer. The nominal wavelength provided by the laser source manufacturer serves as a reference, and the spectrometer monitors the spectral characteristics of the laser beam in real time to obtain actual spectral data. The nominal wavelength is compared with the center wavelength of the actual spectrum to identify wavelength drift or deviation. The center wavelength is determined through spectral peak analysis, and the wavelength value is output. This process verifies the stability of the exposure light source and prevents wavelength fluctuations from affecting the Tiber self-imaging effect.

[0031] The numerical values ​​of the first or second design period are encapsulated with the wavelength values ​​to generate a parameter data package. The data encapsulation uses a standardized format, such as binary or XML structure, integrating the period and wavelength parameters. The parameter data package serves as system input and is transmitted to subsequent calculation modules to ensure data flow consistency. The encapsulation process includes data verification and redundancy checks to prevent invalid parameters from entering the process. The generation of the parameter data package completes the initialization task of step 1, laying the foundation for the mathematical operations in step 2.

[0032] Specifically, in the adaptive lithography method for waveguide master defects based on Tybo self-imaging described in this invention, step 2, which involves extracting design period values ​​and wavelength values ​​from a parameter data package and performing calculations on the design period values ​​and wavelength values ​​to obtain the Tybo self-imaging distance value, includes: Receive parameter data packets and extract the design period value and wavelength value from the parameter data packets; The separated design period values ​​are squared to generate the period square value, and the separated wavelength values ​​are reciprocal to generate the wavelength reciprocal value. Multiplying the squared value of the period by the reciprocal of the wavelength yields a preliminary calculation result for the Talber distance; Using the preliminary calculation result of the Talbot distance as the initial distance parameter, optical diffraction simulation calculation is performed. If the light field uniformity value output by the simulation calculation is lower than the preset standard value, the initial distance parameter is iteratively adjusted until the light field uniformity value reaches the preset standard value, and the final Talbot self-imaging distance value is output.

[0033] The system receives the parameter data packet generated in step 1. This data packet is decapsulated via a data parsing interface, separating the design period value and wavelength value. The design period value may originate from the period markings in the optical waveguide master design drawing or from the raw period signal collected by optical measurement equipment. The wavelength value integrates the nominal wavelength of the laser source and the actual spectral center wavelength monitored by the spectrometer. The separation process employs a data verification algorithm to validate the parameter integrity and prevent invalid data from flowing into subsequent calculation stages.

[0034] The separated design period values ​​are squared to generate the period square value. This square calculation is performed by the arithmetic unit of the embedded processor, amplifying the impact of the period parameter on distance calculation. Simultaneously, the wavelength values ​​are reciprocalized to generate the wavelength reciprocal value. This reciprocal calculation is implemented by the floating-point unit, converting the wavelength unit to reciprocal form to accommodate multiplication requirements. Numerical stability methods are employed during the calculation process to avoid division-by-zero errors or overflow.

[0035] The squared value of the period and the reciprocal of the wavelength are input into a multiplier circuit, multiplied, and the output is a preliminary calculation result of the Talbot distance. The multiplication operation is based on the mathematical relationship of the Talbot self-imaging principle, and the preliminary calculation result represents the theoretical self-imaging position. This result is transmitted as the initial distance parameter to the optical diffraction simulation module.

[0036] The optical diffraction simulation module takes the initial distance parameters as input and performs Fresnel diffraction numerical simulation to calculate the optical field uniformity at the substrate location. The simulation model incorporates the diffraction characteristics of the periodic structure of the master substrate, phase perturbations caused by defects, and the spatial coherence of the light source. The optical field uniformity value is compared with a preset standard value. If it is lower than the standard, the simulation module adjusts the initial distance parameters through an iterative algorithm, such as using gradient descent to optimize the distance value. The adjustment process is repeated multiple times based on the feedback of the optical field distribution until the optical field uniformity value reaches the preset standard. Finally, the Tiber self-imaging distance value is output for substrate positioning control in step 3.

[0037] Each step is connected via data flow, parameters are separated to provide input for calculations, mathematical operations are transformed into physical distance parameters, and simulation verification enhances the reliability of practical applications. The logical relationship is reflected in the data-driven workflow, forming a closed loop from parameter acquisition to distance optimization, ensuring the adaptability of the Taibo self-imaging distance value in the photolithography process.

[0038] Specifically, in the adaptive photolithography method for waveguide master plate defects based on Tybo self-imaging described in this invention, step 3, which involves converting the Tybo self-imaging distance value into a displacement control signal to drive the photoresist-coated substrate to move, and determining that positioning is complete when the difference between the measured distance between the substrate and the master plate and the Tybo self-imaging distance value is less than a set threshold, includes: Read the self-imaging distance value of the Titan and generate a displacement control signal based on the self-imaging distance value of the Titan; The displacement control signal controls the substrate to move towards the optical waveguide mother plate; During the movement, the real-time distance between the base surface and the mother plate surface is measured to obtain the real-time distance measurement value; Calculate the absolute difference between the real-time distance measurement value and the Taibo self-imaging distance value, and determine whether the absolute difference is less than a set threshold. If the absolute difference is greater than or equal to the set threshold, a feedback control signal is generated based on the sign of the difference. The feedback control signal is superimposed on the displacement control signal, so that the value of the displacement control signal is adjusted according to the difference ratio, guiding the base to continue moving. If the absolute difference is less than the set threshold, the feedback control signal will stop being generated and a positioning completion status signal will be output.

[0039] The system reads the TIG self-imaging distance value and generates a displacement control signal based on it. This displacement control signal is converted into an analog voltage signal via a digital-to-analog converter, driving a linear actuator or piezoelectric ceramic platform to move the photoresist-coated substrate vertically towards the surface of the optical waveguide master plate. During this movement, a high-precision displacement sensor, such as a laser interferometer or a capacitance micrometer, continuously monitors the gap between the substrate surface and the master plate surface, outputting real-time distance measurements.

[0040] A difference calculation module calculates the absolute difference between the real-time distance measurement value and the Tiber self-imaging distance value. A comparator compares the absolute difference with a set threshold; this threshold is pre-configured based on the lithography system's accuracy requirements, such as wavelength fraction or platform resolution. If the absolute difference is greater than or equal to the set threshold, the feedback control module generates a feedback control signal based on the sign of the difference. This feedback control signal adjusts the amplitude of the displacement control signal using a proportional-integral-differential algorithm, guiding the substrate to continue moving to reduce gap error.

[0041] If the absolute difference is less than the set threshold, the feedback control module stops outputting the feedback control signal, and the displacement control signal maintains its current value to keep the substrate stationary. Simultaneously, the system outputs a positioning completion status signal, which is a digital level signal, triggering subsequent coherent light output and substrate movement. The positioning process achieves nanometer-level precision through closed-loop control, ensuring a smooth substrate movement trajectory to avoid impact and guaranteeing that the distance between the photoresist layer and the master substrate remains stable at the Tiber self-imaging distance.

[0042] The displacement control signal generation stage is integrated into the system's main controller, and the Tiber self-imaging distance value is read from the parameter storage unit. Real-time distance measurements are sampled through the sensor interface circuit, with the sampling frequency adapted to the platform's dynamic response requirements. Feedback control signals are superimposed using analog adder circuits or digital signal processing units to achieve real-time correction. The positioning completion status signal is connected to the exposure control unit as a trigger condition for process synchronization.

[0043] In specific application scenarios, a larger step speed is used in the initial stage of base movement to quickly approach the target position; when the real-time distance measurement value approaches the Tabo self-imaging distance value, it switches to a micro-stepping mode to improve positioning accuracy. Displacement sensor data is processed by Kalman filtering to suppress environmental vibration noise and improve measurement stability. Feedback control parameters are tuned online based on the platform's inertia and damping characteristics to avoid overshoot or oscillation.

[0044] After the substrate positioning completion status signal is output, the system enters the pre-exposure state. The displacement control signal remains active in closed loop, continuously compensating for minute displacements caused by thermal drift or mechanical creep. The positioning process log records key parameters such as the final difference, number of adjustments, and time consumption for process optimization and fault diagnosis.

[0045] Specifically, in the adaptive photolithography method for waveguide master defects based on Tyber self-imaging described in this invention, step 4, which involves activating the coherent light source output after positioning is completed, and simultaneously controlling the substrate to perform periodic in-plane motion with a preset frequency and amplitude, includes: Receive the positioning completion status signal; In response to the positioning completion status signal, coherent light is generated and modulated to form a uniform illumination field; Based on the photosensitive properties of the photoresist and the period of the target grating structure, the substrate motion frequency parameters and amplitude parameters are determined; The base is controlled to perform in-plane reciprocating motion according to frequency and amplitude parameters; During the reciprocating motion, the actual displacement data of the base is monitored, and the actual displacement data is compared with the frequency parameters and amplitude parameters. If the deviation exceeds the tolerance, the amplitude parameter in the motion control parameters is adjusted proportionally according to the magnitude of the deviation.

[0046] Upon completion of positioning, the system receives a positioning completion status signal from step 3 via an interface circuit. This signal serves as a trigger to initiate the exposure sequence. In response to the positioning completion status signal, the exposure control module sends a drive command to the coherent light source, activating the laser to output a coherent beam. The coherent beam then enters the optical modulation unit, where the spatial light modulator performs phase shaping on the beam according to preset wavefront parameters, eliminating light intensity inhomogeneities and forming a uniform illumination field covering the entire master image. The establishment of this uniform illumination field depends on the calibration of the light source divergence angle and coherence length. During the modulation process, the light field distribution is monitored in real time, and the modulator voltage is dynamically adjusted via a feedback loop to keep light intensity fluctuations within acceptable limits.

[0047] Based on the photosensitive properties of the photoresist, including photosensitizer concentration and exposure threshold, as well as the periodic dimensions of the target grating structure, the parameter calculation module calls upon photoresist response curves and grating design specifications from the material database. An interpolation algorithm generates frequency and amplitude parameters for the substrate motion. The frequency parameter is correlated with the critical exposure dose of the photoresist, while the amplitude parameter is set to a value much smaller than the master plate period according to the Rayleigh criterion to avoid image degradation caused by motion. After the motion parameters are determined, the motion controller converts the frequency and amplitude parameters into drive signals, driving a linear actuator to move the substrate in a reciprocating motion within a plane. The motion trajectory uses a sinusoidal waveform to smooth the acceleration changes.

[0048] During the reciprocating motion of the substrate, a grating ruler or laser interferometer integrated on the motion platform continuously collects actual displacement data, which is converted into digital signals by a data acquisition card. The control software compares the actual displacement data with preset frequency and amplitude parameters in real time. When the displacement deviation exceeds the system tolerance, the adaptive control algorithm adjusts the amplitude parameters proportionally to the deviation, and the adjustment is output to the actuator driver via a digital-to-analog converter. This closed-loop control mechanism compensates for the return error and load disturbance of the transmission mechanism, ensuring that the substrate's motion trajectory always matches the exposure requirements, providing a stable motion foundation for the subsequent homogenization of the diffraction field.

[0049] Specifically, in the adaptive photolithography method for optical waveguide master plate defects based on Talbot self-imaging described in this invention, step 5 involves the coherent light passing through the optical waveguide master plate to form a diffracted light field. When the diffracted light field propagates to the substrate position, it cancels the distortion component caused by the defect through the Fresnel diffraction effect, including: The uniform illumination field formed in step 4 is received, and the uniform illumination field illuminates the defective optical waveguide mother plate. The periodic structure and defects on the optical waveguide mother plate jointly modulate the illumination light field, generating a diffracted light field, which includes information about the mother plate structure and defect information. The diffracted light field propagates from the optical waveguide mother plate to the substrate. Multiple acquisition points are set along the propagation path of the diffracted light field, and the phase data and amplitude data of the diffracted light field at the acquisition points are acquired by a photodetector. Perform Fourier transform on the phase data and amplitude data to convert the diffracted light field into a spatial frequency spectrum, and identify the high-frequency components corresponding to the periodic structure of the master plate and the low-frequency components corresponding to the defect distortion from the spatial frequency spectrum. Based on the principle of Taybernating self-imaging, high-frequency components coherently superimpose at the Taybernating self-imaging distance due to periodic matching, increasing the light intensity to more than 1.5 times the original value. Low-frequency components diverge at the Taybernating self-imaging distance due to non-periodic characteristics, decreasing the light intensity to less than 30% of the original value. The enhanced high-frequency components and the attenuated low-frequency components are combined at the substrate location to obtain a uniform light intensity distribution.

[0050] In the adaptive photolithography process for defects in the optical waveguide master plate, step five first receives the uniform illumination light field generated in step four. This uniform illumination light field uses a spatial light modulator to perform wavefront shaping on the coherent beam, eliminating light source inhomogeneities and forming a plane wave illumination covering the entire master plate. The uniform illumination light field illuminates the defective surface of the optical waveguide master plate, and the energy distribution of the light field achieves spatial consistency by calibrating the beam divergence angle and coherence length.

[0051] The periodic grating structure on the optical waveguide master plate, along with local defects, interacts with the incident light field. The periodic structure generates a regular diffracted wavefront, while the defect region introduces phase distortion and amplitude perturbation. During modulation, the diffraction order of the periodic structure and the stray light caused by the defects superimpose to generate a composite diffracted light field containing information about the ideal pattern of the master plate and the defect errors. The diffracted light field propagates from the edge of the master plate towards the substrate, following Fresnel diffraction laws. Multiple sampling points are set along the propagation path, positioned on equidistant planes between the master plate and the substrate.

[0052] Phase and amplitude data of the diffracted light field at the acquisition point are collected using a photodetector array. The photodetector employs a silicon-based CCD or InGaAS sensor to simultaneously capture the spatial distribution of the light field. After analog-to-digital conversion, the acquired data is input into a signal processing unit, where a fast Fourier transform algorithm is performed on the phase and amplitude data to map the spatial diffraction field into a spatial frequency spectrum. In the frequency spectrum, the high-frequency components corresponding to the master period are concentrated in the diffraction order sub-peak region, while the low-frequency components corresponding to defect distortion appear as background noise.

[0053] Based on the Talbot self-imaging principle, high-frequency components exhibit constructive interference at the Talbot self-imaging distance due to periodic matching, resulting in a significant increase in light intensity through coherent superposition. Low-frequency components, due to their aperiodic characteristics, exhibit energy divergence, leading to light intensity attenuation. The synthesis process uses a digital wavefront reconstruction algorithm to vector-superimpose the enhanced high-frequency components and the attenuated low-frequency components on the substrate plane, ultimately generating a uniform light intensity distribution. This optical cancellation mechanism suppresses local intensity fluctuations caused by defects through integral averaging, providing an ideal patterning basis for photoresist exposure.

[0054] In practice, the sampling frequency of the photodetector must match the propagation speed of the diffraction field to avoid aliasing errors; the Fourier transform uses a window function to optimize spectral leakage and improve the accuracy of frequency component separation. A phase compensation algorithm is introduced in the synthesis stage to correct propagation path errors and ensure the spatial consistency of the uniform light intensity distribution.

[0055] Specifically, in the adaptive photolithography method for optical waveguide master defects based on Tyber self-imaging described in this invention, step 6 involves the photoresist recording the light intensity distribution after cancellation processing, followed by treatment with a developing solution to form a grating structure, including: The uniform light intensity distribution obtained in step 5 is propagated to the substrate surface coated with photoresist through an optical system; The uniform light intensity distribution acts on the photoresist layer, causing the photosensitive components in the photoresist to absorb the photon energy in the light intensity distribution; After absorbing photon energy, the photosensitive component undergoes a photochemical reaction, which leads to cross-linking or decomposition of the photosensitive component in the exposed area; Based on the spatial intensity variation of light intensity distribution, a grating pattern is formed in the photoresist layer that is potential on the surface. In the region where the light intensity is higher than the photoresist photosensitive threshold, a complete photochemical reaction occurs, while in the region where the light intensity is lower than the photoresist photosensitive threshold, an incomplete photochemical reaction occurs. The developer is applied to the photoresist surface to dissolve the areas of photoresist that have not undergone a complete photochemical reaction, while preserving the areas of photoresist that have undergone a complete photochemical reaction. After dissolution, a grating structure corresponding to the uniform light intensity distribution is formed on the photoresist surface.

[0056] During the process of recording a uniform light intensity distribution using photoresist, the uniform light intensity distribution output in step five is propagated to the substrate surface coated with photoresist through an optical system. The optical system includes a lens group and a beam expander, used to collimate and uniformly project the light intensity distribution onto the photoresist layer. When the uniform light intensity distribution acts on the photoresist layer, photosensitive components in the photoresist, such as diazonoquinone compounds, absorb photon energy, triggering a photochemical reaction.

[0057] Upon absorbing photon energy, the photosensitive component undergoes molecular structural changes, resulting in cross-linking or decomposition of the photosensitive component in the exposed area. In positive photoresists, the decomposition of the photosensitive component leads to increased solubility; in negative photoresists, the cross-linking of the photosensitive component forms an insoluble network. Spatial variations in light intensity distribution create a grating pattern latent on the surface of the photoresist layer. Regions with light intensity above the photoresist's photosensitivity threshold undergo complete photochemical reactions, resulting in full conversion of the photosensitive component; regions with light intensity below the photoresist's photosensitivity threshold undergo incomplete photochemical reactions, with some photosensitive component retained.

[0058] The developer is applied to the photoresist surface. An alkaline developer, such as tetramethylammonium hydroxide aqueous solution, is used for positive photoresist to dissolve the exposed areas; an organic solvent is used for negative photoresist to dissolve the unexposed areas. The developer dissolves areas of photoresist that have not undergone complete photochemical reaction, while preserving areas that have. After the dissolution process, a grating structure corresponding to the homogenized light intensity distribution is formed on the photoresist surface. The width and spacing of the grating lines are determined by the peak and valley values ​​of the light intensity distribution.

[0059] The grating structure serves as a mask layer, providing a pattern base for subsequent etching processes. The entire process achieves pattern transfer through the local differences in photochemical reactions, and intensity fluctuations caused by defects are suppressed by the homogenized light intensity distribution, thus improving the integrity of the grating structure.

[0060] Specifically, in the adaptive photolithography method for optical waveguide master defects based on Tyber self-imaging described in this invention, step 6, which involves further etching and deposition of the waveguide functional layer to complete the fabrication of the optical waveguide template, includes: The surface morphology of the grating structure formed after development was measured to obtain grating groove depth data, line width data, and period size data. Based on the grating groove depth data, line width data, and period size data, calculate the etching time parameters, etching gas flow rate parameters, and power parameters; Etching is performed according to etching time parameters, etching gas flow rate parameters, and power parameters to transfer the grating pattern on the photoresist surface into the substrate material; Waveguide functional materials are deposited on the substrate surface after pattern transfer, according to preset thickness and refractive index requirements; The optical performance of the template after deposition of waveguide functional material is tested, and the transmission loss and diffraction efficiency are measured. The measurement results are compared with the performance of the template prepared by the defect-free master, and the defect suppression effect evaluation data are output.

[0061] The surface morphology of the grating structure formed after development was measured using an atomic force microscope or scanning electron microscope to scan the grating region and acquire surface height distribution and contour data. The measurement process focused on three key parameters: grating groove depth, linewidth, and period size. Groove depth data was calculated from the vertical height difference, linewidth data was extracted from the horizontal edge contour, and period size was obtained by statistically analyzing the spacing between repeating cells. The measurement results were converted into digital signals and input into the processing system to provide quantitative basis for subsequent processes.

[0062] Based on grating trench depth data, linewidth data, and period size data, an etching parameter calculation algorithm is executed. The calculation process references a material removal rate model, dividing the difference between the target trench depth and the actual measured value by the etching selectivity ratio to derive the etching time parameter. Linewidth data is used to adjust the etching gas flow rate parameter; the flow rate value is determined through gas dynamics simulation to control the sidewall morphology. Period size data, combined with the thermal conductivity of the substrate material, is used to calculate the plasma power parameter, ensuring etching uniformity. The parameter calculation module outputs the set values ​​for etching time, gas flow rate, and power, which are then transmitted to the etching equipment controller.

[0063] The dry etching process is executed according to etching time, etching gas flow rate, and power parameters. The etching equipment starts with an RF power supply to generate plasma. The reactive gas dissociates into active particles under the influence of an electric field, bombarding the substrate surface under the photoresist mask. The etching time parameter controls the process duration, the gas flow rate parameter adjusts the reactive gas injection rate, and the power parameter maintains the plasma density. The reflectance spectrum of the substrate surface is monitored in real time during etching. Etching is terminated when the grating pattern is completely transferred to the substrate material, forming a relief structure with a preset geometry.

[0064] Waveguide functional materials are deposited on the substrate surface after pattern transfer using chemical vapor deposition. The deposition process controls the reactive gas flow rate and deposition time according to thickness requirements, and film growth is monitored in real time using a quartz crystal microbalance. Refractive index requirements are achieved by adjusting the deposition temperature or gas composition, for example, by incorporating specific elements to adjust the material's optical constants. The deposited waveguide functional layer covers a grating structure, forming the core light-guiding region of the optical waveguide template.

[0065] Optical performance tests were performed on the template after deposition of waveguide functional materials. Transmission loss was measured using a fiber-coupled light source and a spectrometer. Diffraction efficiency was measured by illuminating the grating with an incident laser beam and collecting diffraction order intensities to calculate the efficiency value. The measurement results were compared with data from a reference template prepared from a defect-free master template. The differences in transmission loss and diffraction efficiency were analyzed. The evaluation data output was a defect suppression effect report, used to verify the effectiveness of the photolithography method.

[0066] Specifically, the adaptive photolithography method for optical waveguide master plate defects based on Tyber self-imaging described in this invention further includes: In step 3, the current distance data between the base and the master plate is obtained in real time; in step 4, the base motion frequency data and amplitude data are obtained in real time. Establish the correspondence between distance data and motion frequency data and amplitude data; When the difference between the current distance data and the Tiber self-imaging distance value is less than twice the set threshold, the motion amplitude data will be reduced to 50% of its original value. When the base motion trajectory deviates, adjust the feedback control signal in step 3; The generation of the positioning completion status signal simultaneously satisfies both the distance threshold condition and the motion stability condition.

[0067] A photoresist layer is coated onto the substrate surface using a spin-coating process to form a uniform thin film. The film thickness matches the design requirements of the optical waveguide template. A uniform light intensity distribution is projected onto the photoresist surface through an optical propagation path. The spatial modulation of the light intensity distribution originates from the Talbot self-imaging effect, and the light intensity profile received by the photoresist layer reflects the idealized distribution of the master template's periodic structure. After absorbing photon energy, the photosensitive components in the photoresist undergo photochemical reactions. When the light intensity in the exposed area exceeds the photoresist's photosensitivity threshold, photosensitive components such as diazonoquinone compounds decompose in positive photoresist, or cinnamate esters undergo cross-linking polymerization in negative photoresist. The spatial variation in light intensity distribution leads to differences in energy absorption within the photoresist layer. In high-intensity regions, the photochemical reaction proceeds fully, resulting in a complete transformation of the photosensitive molecular structure. In low-intensity regions, the reaction is insufficient, and the photosensitive components remain in their initial state.

[0068] Spatial modulation of light intensity distribution forms a latent grating pattern within the photoresist layer, the morphology of which corresponds to the intensity gradient of the homogenized light intensity distribution. High-intensity areas form exposed zones, where the photoresist's chemical properties change; low-intensity areas form unexposed zones, where the photoresist retains its original chemical characteristics. A developer is coated onto the photoresist surface; the developer composition is selected based on the photoresist type. Alkaline solutions, such as tetramethylammonium hydroxide aqueous solution, are used for positive photoresists, while acidic or organic solvents are used for negative photoresists. The developer reacts chemically with the photoresist, dissolving areas where the photochemical reaction has not fully occurred. The exposed zones of positive photoresist are dissolved, as are the unexposed zones of negative photoresist. After the developer's action, the fully photochemically reacted areas remain on the substrate surface, forming a grating structure with a physical morphology.

[0069] After dissolution, the photoresist surface exhibits a grating structure corresponding to the uniform light intensity distribution. The periodicity of the grating structure is consistent with the design period of the master template, and the line width and spacing are determined by the peak and valley values ​​of the light intensity distribution. The grating structure serves as a mask layer, providing the patterning basis for subsequent etching processes. The etching process employs either dry or wet etching methods to transfer the pattern from the photoresist surface into the substrate material. Dry etching uses reactive ion etching (RIE) technology, bombarding the substrate surface with plasma, while wet etching uses chemical solutions to selectively etch the substrate material. After pattern transfer, residual photoresist is removed, and waveguide functional layer material is deposited to form the optical waveguide template functional structure. In this process, the photoresist recording and development steps convert optical information into a physical structure, enabling pattern replication after defect suppression. The optical waveguide template is used for the mass production of optical communication devices.

[0070] Specifically, the adaptive photolithography method for optical waveguide master plate defects based on Tyber self-imaging described in this invention further includes: The value of the self-imaging distance of the Titanium laser output in step 2 is used as the basis for generating the displacement control signal in step 3. The positioning completion status signal generated in step 3 serves as the trigger condition for starting exposure in step 4. The substrate motion start time in step 4 is synchronized with the coherent light output start time; Monitor the self-imaging distance value, actual distance measurement value, motion frequency data, motion amplitude data, and light intensity distribution data of the Titanium laser. If any data exceeds the preset normal range threshold, the process is paused and a specific adjustment instruction is generated. The adjustment instruction includes the data name, deviation amount, and adjustment direction. The process continues after the data returns to the normal range threshold.

[0071] In the adaptive lithography method for waveguide master defects based on Tybo self-imaging, the automated control of the signal chain involves the coordinated operation of multiple steps. Seamless integration of data streams and status signals enhances process stability. The Tybo self-imaging distance value calculated in step two is directly transmitted to the displacement control unit as the basic input for generating the displacement control signal. The displacement control unit drives the photoresist-coated substrate to move according to the Tybo self-imaging distance value. During the movement, the distance measurement between the substrate and the master is acquired in real time, and the displacement is dynamically adjusted through a feedback loop until the difference between the actual distance measurement value and the Tybo self-imaging distance value is less than a set threshold.

[0072] Step three, after completing substrate positioning, generates a positioning completion status signal, which serves as the trigger condition for step four, exposure initiation. Upon receiving the positioning completion status signal, the exposure control module activates the coherent light source and simultaneously initiates periodic movement within the substrate plane. The start time of substrate movement and the start time of coherent light output are strictly synchronized via a synchronization circuit or software protocol to prevent uneven light field distribution caused by exposure timing misalignment.

[0073] During automated control, the system continuously monitors the self-imaging distance, actual distance measurement, motion frequency data, motion amplitude data, and light intensity distribution data. The monitoring data is collected via a sensor network and transmitted to the central processing unit (CPU), which compares the real-time data with preset normal range thresholds. When any monitored data exceeds the normal range threshold, the system automatically pauses the lithography process and generates an adjustment command, which includes the data name, deviation amount, and adjustment direction. The adjustment command is then sent to the corresponding actuators, such as the displacement platform or the light source modulator. The process resumes once the monitored data returns to within the normal range threshold.

[0074] This signal chain design achieves closed-loop control from steps two to four. The data flow extends from distance calculation to exposure execution, and the logical relationship is that the output of the preceding step serves as the input for the subsequent step. The positioning completion status signal acts as a key trigger point connecting substrate positioning and exposure initiation, while the data monitoring mechanism ensures the robustness of the signal chain. In practical applications, automated control reduces manual intervention and improves the repeatability and defect suppression effect of the lithography process.

[0075] Step 2: Extract the design period value and wavelength value from the parameter data package, and perform calculations on the design period value and wavelength value to obtain the Tiber self-imaging distance value, including: Receive parameter data packets and extract the design period value and wavelength value from the parameter data packets; Based on the principle of Tiber self-imaging, the theoretical formula for calculating the Tiber self-imaging distance is as follows: ; in, Indicates the self-image distance of the Titan. This indicates the design period value of the optical waveguide motherboard. This indicates the wavelength value of the coherent light used for exposure.

[0076] In the specific calculation, the separated design period values ​​are first squared to generate the period square value. ,in This represents the design period value of the optical waveguide motherboard; the reciprocal of the separated wavelength value is calculated to produce the reciprocal wavelength value. ,in This represents the numerical value of the coherent light wavelength used for exposure; the squared value of the period. Reciprocal of wavelength Multiplying the results, we get the preliminary calculated result of the Timber distance; ; in This indicates the preliminary calculation results of the Timber distance; Due to potential deviations in actual optical systems, the preliminary calculation results based on the Talber distance... As the initial distance parameter, optical diffraction simulation calculations are performed. If the light field uniformity value output by the simulation calculation is lower than the preset standard value, the initial distance parameter is iteratively adjusted until the light field uniformity value reaches the preset standard value, and the final Tiger self-imaging distance value is output. ,in This represents the final Tiber self-imaging distance value.

[0077] The data processing path is as follows: design cycle values ​​extracted from the parameter data package. and wavelength values As input, the intermediate variable periodic square value is obtained by successively performing square and reciprocal calculations. and the reciprocal of wavelength Then, the preliminary calculation results of the Timber distance are obtained through multiplication operations. Subsequently The initial distance parameter is input into the optical diffraction simulation module, and after iterative adjustments, the final Tiger self-imaging distance value is output. The data processing path is implemented through embedded algorithms or dedicated software to ensure the accuracy and reliability of the calculation results.

[0078] In practical applications of optical waveguide template fabrication, micron- or nanometer-scale defects, such as scratches or periodic disturbances, that may exist on the surface of the master template can be directly transmitted to the photoresist template, leading to performance degradation in the final device. This invention provides an adaptive lithography method that compensates for these defects through the Tiber self-imaging effect, eliminating the need for physical repair of the master template. The core of the method lies in precisely controlling the relative position and motion state between the photoresist substrate and the master template, utilizing the self-imaging averaging effect during Fresnel diffraction to achieve optical self-healing of defect perturbations on the imaging plane.

[0079] During implementation, the design period parameters of the optical waveguide master plate and the wavelength parameters of the coherent light used for exposure are first acquired. The design period value can be directly read from the master plate design drawings or obtained by scanning the master plate surface with high-precision optical measurement equipment to obtain actual period data; the wavelength parameters are determined in real time by combining the laser source output specifications with spectrometer monitoring. The basic parameters are encapsulated into data packets to provide input for subsequent calculations. Accuracy is emphasized in the parameter acquisition process; for example, digital filtering is used to process the raw signal to eliminate noise and ensure the reliability of subsequent steps.

[0080] Next, based on the acquired period and wavelength data, the Talbot self-imaging distance is calculated. The calculation process follows the Talbot effect principle: the distance is directly proportional to the square of the period and inversely proportional to the wavelength. In actual calculations, an embedded algorithm or dedicated software is used to perform the sum of squares and reciprocal calculations to generate preliminary distance values. For further optimization, optical diffraction simulation can be introduced, using the preliminary results as a starting point for iterative adjustments until the optical field uniformity reaches a preset standard. This step ensures that the photoresist is located on the optimal imaging plane, creating a prerequisite for the defect averaging effect.

[0081] The calculated Tbauer distance is converted into a displacement control signal, driving the photoresist-coated substrate to move. The displacement control signal guides the substrate towards the master substrate, while a high-precision sensor monitors the gap between the substrate and the master substrate in real time. When the difference between the real-time distance measurement and the Tbauer distance is less than a set threshold, the system determines that positioning is complete. The movement process employs closed-loop control, dynamically adjusting the displacement through a feedback mechanism to avoid overshoot or oscillation. For example, a piezoelectric ceramic platform can achieve nanometer-level positioning, ensuring the substrate remains stably at the Tbauer distance.

[0082] After positioning, a coherent light source is activated to output a uniform illumination field, and the substrate is simultaneously controlled to perform periodic micro-motions within the plane. The motion parameters are set based on the photoresist's photosensitive characteristics and the target grating structure period, with the amplitude much smaller than the master plate period to prevent image blurring. The motion introduces a spatial averaging effect, diluting defect disturbances during exposure in the time domain. The motion control module monitors the substrate's displacement trajectory in real time; if the deviation exceeds the tolerance, it adaptively adjusts the amplitude to maintain motion stability.

[0083] Coherent light undergoes Fresnel diffraction after passing through a defective master substrate. The diffracted light field propagates to the substrate and is then integrated and averaged over the substrate surface. The periodic diffracted light from intact regions reinforces each other, forming stable interference fringes; stray light from defects has low energy and is overwhelmed by surrounding signals during integration. The light field distribution data is acquired using a multi-channel sensor, and Fourier transform is used to separate high-frequency and low-frequency components. At the Talbot distance, the high-frequency components are enhanced due to periodic matching, while the low-frequency components diverge, thus achieving optical cancellation of defect distortion.

[0084] The photoresist records the averaged light intensity distribution, and its photosensitive components selectively react according to the light intensity. After development, only the complete grating structure formed by the diffracted light from the intact areas is retained, while the defective areas are dissolved because the light intensity is insufficient to trigger exposure. Subsequently, the pattern is transferred to the substrate material through etching, and a waveguide functional layer is deposited, ultimately obtaining a defect-free optical waveguide template. The process is coordinated by an automated signal chain, such as using Talbot distance data to directly drive displacement control, triggering exposure upon positioning completion, and real-time monitoring of various parameters to ensure process stability. This method effectively suppresses defect propagation, reduces master template scrap rate and manufacturing costs, and is suitable for the large-scale production of high-precision optical waveguide devices.

[0085] In practical applications of optical waveguide template fabrication, scratches on the master template surface can lead to corresponding grooves, affecting the integrity of the grating structure. This embodiment addresses this defect by implementing an adaptive photolithography method. First, the grating period markings are read from the optical waveguide master template design drawings, yielding a design period value of 650 nm. Simultaneously, a 355 nm ultraviolet laser is used as the exposure source, and the output stability is calibrated using a spectrometer. The parameter data package is encapsulated and input into the calculation module, and the theoretical imaging distance of 2.38 mm is derived based on the Talbot self-imaging principle. The silicon substrate coated with positive photoresist is precisely positioned at this distance using a piezoelectric ceramic platform. During the positioning process, a laser interferometer monitors the gap distance in real time, and a positioning completion signal is triggered when the measured value deviates from the theoretical value by less than 50 nm.

[0086] After coherent light output, the substrate undergoes in-plane sinusoidal motion at a frequency of 200 Hz and an amplitude of 5 nm. During this motion, stray diffracted light from the defect region is averaged by the strong signal from the surrounding intact region during Fresnel diffraction integration. After photoresist exposure, a 2.38% tetramethylammonium hydroxide solution is used for development to form a grating structure with a linewidth of 325 nm. Subsequently, the pattern is transferred to the silicon substrate via reactive ion etching. After depositing a silicon dioxide waveguide layer, the template diffraction efficiency is measured to be 81%, with a performance deviation of less than 3% compared to the template prepared from a defect-free master.

[0087] For master substrates with localized periodic disturbances, this embodiment uses actual measured periods instead of design values. Periodic data from multiple regions on the master substrate surface were collected using a scanning electron microscope, and the average period was obtained as 632 nm after digital filtering. Combined with a 325 nm wavelength helium-cadmium laser, the Talbot distance was calculated to be 1.96 mm. During the substrate positioning stage, an adaptive algorithm was used to dynamically adjust the threshold; positioning stability was determined when the substrate vibration amplitude was less than 0.5 nm. During exposure, the substrate moved at a frequency of 150 Hz and an amplitude of 3 nm, effectively suppressing diffraction distortion caused by abrupt periodic changes.

[0088] Atomic force microscopy (AFM) analysis of the developed grating structure showed a 40% improvement in linewidth uniformity in the periodically disordered regions compared to existing methods. In template transmission loss testing, a loss value of 0.8 dB / cm was measured in the 1550 nm communication band, showing no significant difference compared to the reference template. This embodiment validates the method's ability to compensate for periodic defects and is suitable for fabricating highly uniform arrayed waveguide grating devices.

[0089] During implementation, key parameters such as the motion frequency need to be adjusted according to the photosensitivity of the photoresist. For high-sensitivity chemically amplified photoresists, the motion frequency can be increased to 500 Hz to enhance the time averaging effect; while for thick-film photoresists, the frequency needs to be reduced to 50 Hz to ensure uniform exposure depth. The amplitude setting always follows the Rayleigh criterion and is controlled below 1 / 10 of the master cycle to avoid image blurring. This parameter adaptive mechanism allows the method to flexibly respond to different process requirements, maintaining the tolerance of the process window while ensuring defect suppression.

Claims

1. An adaptive photolithography method for defects in optical waveguide master plates based on Tyber self-imaging, characterized in that, include: Step 1: Collect the design cycle value of the optical waveguide master and the coherent light wavelength value for exposure, and generate a parameter data package; Step 2: Extract the design cycle value and wavelength value from the parameter data package, and perform calculations on the design cycle value and wavelength value to obtain the Tiber self-imaging distance value. Step 3: Convert the Talbot self-imaging distance value into a displacement control signal to drive the substrate coated with photoresist to move. When the difference between the measured distance between the substrate and the master plate and the Talbot self-imaging distance value is less than the set threshold, the positioning is determined to be complete. Step 4: Once the positioning is complete, start the coherent light output and simultaneously control the substrate to perform periodic in-plane motion with preset frequency and amplitude parameters. Step 5: Coherent light passes through the optical waveguide mother plate to form a diffracted light field. During the propagation of the diffracted light field to the substrate position, the distortion component caused by the defect is canceled by the Fresnel diffraction effect. The distortion component is the non-uniform part of the diffracted light field caused by the defect. Step 6: The photoresist records the light intensity distribution after the cancellation process, and the photoresist is treated with a developer to form a grating structure. Then, the optical waveguide template is prepared by etching and deposition of waveguide functional layers.

2. The adaptive photolithography method for defects in optical waveguide master plates based on Talbot self-imaging according to claim 1, characterized in that, Step 1 includes: Read the period markings in the optical waveguide master design drawing to generate the first design period value, or use an optical measurement device to scan the surface of the master, collect the original period signal, perform digital filtering on the original period signal, and output the second design period value. The nominal wavelength value of the laser source is collected, and the actual spectrum of the laser output is monitored using a spectrometer. The nominal wavelength value is compared with the center wavelength of the actual spectrum, and the wavelength value is output. The first or second design period value is encapsulated with the wavelength value to generate a parameter data package.

3. The adaptive photolithography method for defects in optical waveguide master plates based on Tyber self-imaging as described in claim 2, characterized in that, Step 2 includes: Receive parameter data packets and extract the design period value and wavelength value from the parameter data packets; The separated design period values ​​are squared to generate the period square value, and the separated wavelength values ​​are reciprocal to generate the wavelength reciprocal value. Multiplying the squared value of the period by the reciprocal of the wavelength yields a preliminary calculation result for the Talber distance; Using the preliminary calculation result of the Talbot distance as the initial distance parameter, optical diffraction simulation calculation is performed. If the light field uniformity value output by the simulation calculation is lower than the preset standard value, the initial distance parameter is iteratively adjusted until the light field uniformity value reaches the preset standard value, and the final Talbot self-imaging distance value is output.

4. The adaptive photolithography method for defects in optical waveguide master plates based on Tyber self-imaging as described in claim 3, characterized in that, Step 3 includes: Read the self-imaging distance value of the Titanium, generate a displacement control signal based on the self-imaging distance value of the Titanium, and control the substrate to move towards the optical waveguide mother plate using the displacement control signal. Measure the real-time distance between the substrate surface and the master plate surface to obtain the real-time distance measurement value; Calculate the absolute difference between the real-time distance measurement value and the Taibo self-imaging distance value, and determine whether the absolute difference is less than a set threshold. If the absolute difference is greater than or equal to the set threshold, a feedback control signal is generated based on the sign of the difference. The feedback control signal is superimposed on the displacement control signal, so that the value of the displacement control signal is adjusted according to the difference ratio, guiding the base to continue moving. If the absolute difference is less than the set threshold, the feedback control signal will stop being generated and a positioning completion status signal will be output.

5. The adaptive photolithography method for defects in optical waveguide master plates based on Titan self-imaging according to claim 4, characterized in that, Step 4 includes: Receive the positioning completion status signal; In response to the positioning completion status signal, coherent light is generated and modulated to form a uniform illumination field; Based on the photosensitive properties of the photoresist and the period of the target grating structure, the substrate motion frequency parameters and amplitude parameters are determined; The base is controlled to perform in-plane reciprocating motion according to frequency and amplitude parameters; During the reciprocating motion, the actual displacement data of the base is monitored, and the actual displacement data is compared with the frequency parameters and amplitude parameters. If the deviation exceeds the tolerance, the amplitude parameter in the motion control parameters is adjusted proportionally according to the magnitude of the deviation.

6. The adaptive photolithography method for defects in optical waveguide master plates based on Tyber self-imaging according to claim 5, characterized in that, Step 5 includes: The uniform illumination field formed in step 4 is received, and the uniform illumination field illuminates the defective optical waveguide mother plate. The periodic structure and defects on the optical waveguide mother plate jointly modulate the illumination light field, generating a diffracted light field, which includes information about the mother plate structure and defect information. The diffracted light field propagates from the optical waveguide mother plate to the substrate. Multiple acquisition points are set along the propagation path of the diffracted light field, and the phase data and amplitude data of the diffracted light field at the acquisition points are acquired by a photodetector. Perform Fourier transform on the phase data and amplitude data to convert the diffracted light field into a spatial frequency spectrum, and identify the high-frequency components corresponding to the periodic structure of the master plate and the low-frequency components corresponding to the defect distortion from the spatial frequency spectrum. Based on the principle of Taybernating self-imaging, high-frequency components coherently superimpose at the Taybernating self-imaging distance due to periodic matching, increasing the light intensity to more than 1.5 times the original value. Low-frequency components diverge at the Taybernating self-imaging distance due to non-periodic characteristics, decreasing the light intensity to less than 30% of the original value. The enhanced high-frequency components and the attenuated low-frequency components are combined at the substrate location to obtain a uniform light intensity distribution.

7. The adaptive photolithography method for defects in optical waveguide master plates based on Talbot self-imaging according to claim 6, characterized in that, In step 6, the photoresist records the light intensity distribution after cancellation processing, and then forms a grating structure after being treated with a developing solution, including: The uniform light intensity distribution obtained in step 5 is propagated to the substrate surface coated with photoresist through an optical system; The uniform light intensity distribution acts on the photoresist layer, causing the photosensitive components in the photoresist to absorb the photon energy in the light intensity distribution; After absorbing photon energy, the photosensitive component undergoes a photochemical reaction, which leads to cross-linking or decomposition of the photosensitive component in the exposed area; Based on the spatial intensity variation of light intensity distribution, a grating pattern is formed in the photoresist layer that is potential on the surface. In the region where the light intensity is higher than the photoresist photosensitive threshold, a complete photochemical reaction occurs, while in the region where the light intensity is lower than the photoresist photosensitive threshold, an incomplete photochemical reaction occurs. The developer is applied to the photoresist surface to dissolve the areas of photoresist that have not undergone a complete photochemical reaction, while preserving the areas of photoresist that have undergone a complete photochemical reaction. After dissolution, a grating structure corresponding to the uniform light intensity distribution is formed on the photoresist surface.

8. The adaptive photolithography method for defects in optical waveguide master plates based on Talbot self-imaging according to claim 7, characterized in that, In step 6, the optical waveguide template is fabricated by further etching and depositing the waveguide functional layer, including: The surface morphology of the grating structure formed after development was measured to obtain grating groove depth data, line width data, and period size data. Based on the grating groove depth data, line width data, and period size data, calculate the etching time parameters, etching gas flow rate parameters, and power parameters; Etching is performed according to etching time parameters, etching gas flow rate parameters, and power parameters to transfer the grating pattern on the photoresist surface into the substrate material; Waveguide functional materials are deposited on the substrate surface after pattern transfer, according to preset thickness and refractive index requirements; The optical performance of the template after deposition of waveguide functional material is tested, and the transmission loss and diffraction efficiency are measured. The measurement results are compared with the performance of the template prepared by the defect-free master, and the defect suppression effect evaluation data are output.

9. The adaptive photolithography method for defects in optical waveguide master plates based on Talbot self-imaging according to claim 8, characterized in that, Also includes: In step 3, the current distance data between the base and the master plate is obtained in real time; in step 4, the base motion frequency data and amplitude data are obtained in real time. Establish the correspondence between distance data and motion frequency data and amplitude data; When the difference between the current distance data and the Tiber self-imaging distance value is less than twice the set threshold, the motion amplitude data will be reduced to 50% of its original value. When the base motion trajectory deviates, adjust the feedback control signal in step 3; The generation of the positioning completion status signal simultaneously satisfies both the distance threshold condition and the motion stability condition.

10. The adaptive photolithography method for defects in optical waveguide master plates based on Talbot self-imaging according to claim 9, characterized in that, Also includes: The value of the self-imaging distance of the Titanium laser output in step 2 is used as the basis for generating the displacement control signal in step 3. The positioning completion status signal generated in step 3 serves as the trigger condition for starting exposure in step 4. The substrate motion start time in step 4 is synchronized with the coherent light output start time; Monitor the self-imaging distance value, actual distance measurement value, motion frequency data, motion amplitude data, and light intensity distribution data of the Titanium laser. If any data exceeds the preset normal range threshold, the process is paused and a specific adjustment instruction is generated. The adjustment instruction includes the data name, deviation amount, and adjustment direction. The process continues after the data returns to the normal range threshold.

Images