An AR / VR resin wafer process

Abstract

The present invention discloses an AR / VR resin wafer process, including: preparing a resin wafer to be tested and a reference test piece, performing positioning and marking on the micro-nano structure area of the resin wafer to be tested to obtain a detection sample; performing multi-dimensional spectral characterization on the detection sample and the reference test piece, determining the material non-uniform distribution area of the detection sample through comparative analysis, and obtaining molecular structure characterization data; performing stress field detection and dynamic response analysis on the material non-uniform distribution area to obtain mechanical property characterization data and high-risk defect points; performing microscopic structure and morphology analysis on the high-risk defect points to obtain defect characteristic data; performing systematic evaluation based on the molecular structure characterization data, mechanical property characterization data and defect characteristic data to obtain the health assessment result of the resin wafer. The technical solution of the present invention can perform multi-scale feature characterization on the micro-nano structure area in the resin wafer for AR / VR, and realizes systematic detection of features such as material molecular state, interface stress and phase change.

Description

Technical Field

[0001] The present invention relates to the technical field of material detection for AR / VR optical devices, and particularly to an ARVR resin wafer process. Background Art

[0002] In the development process of augmented reality (AR) and virtual reality (VR) technologies, resin wafers, as a key component of a new type of optical device, are gradually replacing traditional glass or silicon-based substrate materials due to their advantages such as light weight, high plasticity, and easy micro-nano processing. Such resin wafers are an indispensable core component in AR / VR optical systems, and usually integrated with complex micro-nano structures such as diffraction gratings, waveguide layers, and microlens arrays on them. These structures work together to achieve the optical functions required by AR / VR devices, such as light coupling, conduction, and emission, thereby ensuring that the device has excellent optical properties such as a large field of view, high transmittance, and low distortion.

[0003] However, different from single crystals or homogeneous glasses, resin wafers will undergo significant molecular structure evolution during multiple processes such as curing, etching, and bonding: the internal cross-linked network may form different degrees of chain segment orientation in different regions, and inorganic or organic modified additives will also be dispersed in a non-uniform manner, thus accumulating local stress near the micro-nano structures. More complexly, when these resin wafers are stacked or combined into multi-layer optical elements, the bonding interfaces between layers may exhibit phenomena of both chemical incompatibility and physical stress concentration, leading to the generation of micro-cracks or phase change regions. Such subtle changes at the material level are often not detected in conventional appearance inspections or simple optical transmittance measurements, but will cause a significant deterioration of optical performance through fluctuations in refractive index or birefringence during actual use, thereby affecting the imaging quality and stability of AR / VR devices. Existing material characterization methods mostly focus on measuring homogeneous samples without structural interference, such as conventional infrared spectroscopy, differential scanning calorimetry, stress polarization detection, etc. These methods are often ineffective when faced with resin wafers with complex micro-nano structures, high-resolution waveguides, or diffraction gratings. This is mainly because the etching of micro-nano structures or the diffraction pattern regions themselves generate multiple couplings of deformation and stress, and at the same time, the laminated interfaces of multi-layer composites are not necessarily uniform and flat in the actual functional regions, showing more irregular transitions at the microscopic scale, while conventional material detection cannot provide precise quantification of these regions. Therefore, there is an urgent need to establish a complete detection method system that can simultaneously characterize multi-scale features such as the molecular state, interface stress, and phase change in the micro-nano structure regions of resin wafers for AR / VR. Summary of the Invention

[0004] The main objective of the present invention is to solve the technical problem that existing material characterization methods cannot accurately characterize multi-scale features such as the molecular state, interfacial stress, and phase change of materials when detecting resin wafers for AR / VR with complex micro-nano structures.

[0005] The first aspect of the present invention provides an AR / VR resin wafer process, and the AR / VR resin wafer process includes:

[0006] Prepare a resin wafer to be tested and a reference specimen, locate and mark the micro-nano structure area of the resin wafer to be tested to obtain a test sample;

[0007] Perform multi-dimensional spectral characterization on the test sample and the reference specimen, determine the material non-uniform distribution area of the test sample through comparative analysis, and obtain molecular structure characterization data;

[0008] Perform stress field detection and dynamic response analysis on the material non-uniform distribution area to obtain mechanical property characterization data and high-risk defect points;

[0009] Perform microscopic structure and morphology analysis on the high-risk defect points to obtain defect feature data;

[0010] Conduct a systematic evaluation based on the molecular structure characterization data, mechanical property characterization data, and defect feature data to obtain the health assessment result of the resin wafer.

[0011] Preferably, the step of locating and marking the micro-nano structure area of the resin wafer to be tested to obtain a test sample includes:

[0012] Dust removal, solvent cleaning, and electrostatic treatment are performed on the resin wafer to be tested, and a thermal and humid environment balance treatment is carried out to obtain a pre-treated wafer;

[0013] Obtain the coordinate positions of the diffraction grating edge, bonding layer interface, and micro-lens array center on the pre-treated wafer, and establish a detection path;

[0014] Determine reference points in the micro-nano structure area of the pre-treated wafer according to the detection path, and mark the reference points to obtain a test sample.

[0015] Preferably, the step of obtaining the coordinate positions of the diffraction grating edge, bonding layer interface, and micro-lens array center on the pre-treated wafer and establishing a detection path includes:

[0016] Perform light intensity distribution detection on the diffraction grating of the pre-treated wafer to obtain an optical coupling efficiency gradient map, determine the high-low efficiency demarcation point of the diffraction grating according to the optical coupling efficiency gradient map, and use the high-low efficiency demarcation point as the first coordinate group of the diffraction grating edge;

[0017] Perform focus distribution detection on the microlens array of the preprocessed wafer to obtain a focus offset distribution map, determine the focus distortion demarcation point of the microlens according to the focus offset distribution map, and use the focus distortion demarcation point as the second coordinate group of the center of the microlens array;

[0018] Perform thickness uniformity detection on the bonding layer of the preprocessed wafer to obtain a thickness change rate distribution map, determine the high and low stress demarcation point of the bonding layer according to the thickness change rate distribution map, and use the high and low stress demarcation point as the third coordinate group of the bonding layer interface;

[0019] Assign weights to the first coordinate group, the second coordinate group and the third coordinate group, establish coordinate priorities, and generate a detection node sequence;

[0020] Generate a detection path according to the detection node sequence according to the minimum path algorithm and the maximum coverage principle.

[0021] Preferably, perform multi-dimensional spectral characterization on the test sample and the reference specimen, determine the material non-uniform distribution area of the test sample through comparative analysis, and obtain molecular structure characterization data, including:

[0022] Perform Raman spectroscopy detection and infrared spectroscopy detection on the test sample to obtain first spectral data;

[0023] Perform fluorescence lifetime detection and optical anisotropy detection on the test sample to obtain second spectral data;

[0024] Perform Raman spectroscopy detection, infrared spectroscopy detection, fluorescence lifetime detection and optical anisotropy detection on the reference specimen to obtain reference spectral data;

[0025] Compare and analyze the first spectral data and the second spectral data with the reference spectral data respectively to determine the position coordinates of the material non-uniform distribution area and the molecular structure characterization data.

[0026] Preferably, performing fluorescence lifetime detection and optical anisotropy detection on the test sample to obtain second spectral data includes:

[0027] Apply probe light at different angles to the test sample to obtain multi-angle fluorescence decay curves, calculate the molecular orientation angle distribution according to the fluorescence decay curves, and obtain a molecular chain orientation state map;

[0028] Perform polarization state measurement on the test sample at different incident angles to obtain a polarization degree distribution map, calculate the interface stress-induced birefringence value according to the polarization degree distribution map, and obtain a stress-optical coupling coefficient;

[0029] Scan the waveguide layer of the detection sample to obtain a local refractive index fluctuation map, calculate the optical path difference distribution according to the refractive index fluctuation map and the stress-optical coupling coefficient, and obtain phase delay data;

[0030] Measure the transmittance of the detection sample at different field of view angles to obtain an angle-dependent transmittance curve, and calculate the optical uniformity index according to the angle-dependent transmittance curve and the phase delay data;

[0031] Perform correlation analysis on the molecular chain orientation state map, the stress-optical coupling coefficient, the phase delay data, and the optical uniformity index to obtain second spectral data.

[0032] Preferably, the stress field detection and dynamic response analysis of the material non-uniform distribution area to obtain mechanical property characterization data and high-risk defect points include:

[0033] Perform polarized photoelastic detection on the material non-uniform distribution area to obtain a stress distribution map;

[0034] Perform local mechanical detection on the stress concentration points in the stress distribution map to obtain micro-area mechanical data;

[0035] Apply cyclic stress and temperature changes to the stress concentration points to obtain dynamic response characteristics;

[0036] Perform fatigue characteristic analysis according to the dynamic response characteristics to obtain fatigue characteristic data;

[0037] Integrate and analyze the micro-area mechanical data and the fatigue characteristic data to obtain mechanical property characterization data;

[0038] Determine the high-risk defect points in the stress concentration points according to the mechanical property characterization data.

[0039] Preferably, applying cyclic stress and temperature changes to the stress concentration points to obtain dynamic response characteristics includes:

[0040] Obtain a multi-directional stress loading spectrum according to the ARVR wearing posture simulation, apply the multi-directional stress loading spectrum to the stress concentration points, and obtain a multi-dimensional stress response curve;

[0041] Establish a temperature-humidity combined action cycle according to the actual use conditions of ARVR, apply the temperature-humidity combined action cycle to the stress concentration points, and obtain an environmental response curve;

[0042] Perform Fourier analysis on the multi-dimensional stress response curve to obtain a stress spectral density function, and calculate the cumulative damage rate according to the stress spectral density function;

[0043] Segment the environmental response curve according to different functional regions to obtain regional sensitivity coefficients, and determine the critical failure region based on the regional sensitivity coefficients;

[0044] Obtain the dynamic response characteristics according to the corresponding relationship between the cumulative damage rate and the critical failure region.

[0045] Preferably, perform microstructure and morphology analysis on the high-risk defect points to obtain defect characteristic data, including:

[0046] Perform ultrasonic testing on the high-risk defect points to analyze whether there are voids, delaminations or cracks, and obtain ultrasonic imaging signals;

[0047] Precisely cut the high-risk defect points according to the abnormal echo positions in the ultrasonic imaging signals to obtain nanoscale thin film samples;

[0048] Perform interface morphology analysis on the nanoscale thin film samples to obtain material interface structure data;

[0049] Perform near-field optical detection on the high-risk defect points to obtain refractive index gradient and birefringence distribution data;

[0050] Obtain defect characteristic data based on the material interface structure data, the refractive index gradient and the birefringence distribution data.

[0051] Preferably, perform near-field optical detection on the high-risk defect points to obtain refractive index gradient and birefringence distribution data, including:

[0052] Scan and image the high-risk defect points at different field angles to obtain the curve of the change of the point spread function with the angle, and calculate the imaging quality attenuation coefficient according to the change curve;

[0053] Measure the optical field distribution of the waveguide structure around the high-risk defect points to obtain a near-field light intensity distribution map, and analyze the optical conduction loss characteristics according to the light intensity distribution map;

[0054] Establish a defect influence radius according to the imaging quality attenuation coefficient, and perform a fine scan of the refractive index within the influence radius to obtain a high-resolution refractive index gradient map;

[0055] Perform wavelet transform analysis on the high-resolution refractive index gradient map, extract local characteristic frequencies, and calculate the optical anomaly index according to the characteristic frequencies;

[0056] Comprehensively evaluate the optical conduction loss characteristics and the optical anomaly index to obtain refractive index gradient and birefringence distribution data.

[0057] Preferably, the system evaluation is performed according to the molecular structure characterization data, mechanical property characterization data, and defect feature data to obtain the health assessment result of the resin wafer, including:

[0058] Perform coordinate matching on the molecular structure characterization data, mechanical property characterization data, and defect feature data to obtain the spatial distribution relationship of the multi-source data;

[0059] Perform cross-validation on the multi-source data according to the spatial distribution relationship to determine the intersection region that shows anomalies in different detection dimensions;

[0060] Visualize and render the intersection region in a three-dimensional coordinate system to generate a three-dimensional defect distribution map;

[0061] Calculate the local crosslinking degree deviation, stress concentration coefficient, and defect diffusion coefficient of the intersection region;

[0062] Perform a comprehensive scoring according to the local crosslinking degree deviation, stress concentration coefficient, and defect diffusion coefficient to obtain defect scoring data;

[0063] Perform a grading process on the defect scoring data to obtain the health assessment result of the resin wafer.

[0064] The detection and analysis process of the technical solution provided by the embodiment of the present application can be divided into several main stages. First, it is necessary to prepare a resin wafer and a reference test piece for comparison, and mark the key micro-nano structure regions on the resin wafer. This preparation work not only includes dust removal, cleaning, and environmental balance treatment of the wafer surface, but also involves accurately capturing and recording the coordinate positions of functional areas such as diffraction gratings, bonding layer interfaces, and microlens arrays. Through this step, it is possible to pre-position various places that may generate local stress or material inhomogeneity, making the subsequent detection more targeted. If these key micro-regions are not accurately marked at the beginning, sampling deviation may occur in the subsequent detection, or it may be impossible to summarize the true material property distribution due to excessive structural differences.

[0065] Subsequently, various spectroscopic techniques are needed to comprehensively characterize the labeled samples and control specimens, with particular attention paid to dimensions such as Raman, infrared, fluorescence lifetime, and polarization state. These methods are each sensitive to the molecular bonding patterns, local crosslinking degrees, fluorescence effects, and even optical anisotropy of materials. By comparing the results of the same batch of data with those of the control specimens, the regions where the wafers exhibit abnormalities at the molecular level can be precisely identified. For example, if the Raman peak positions shift in certain microregions, or the fluorescence lifetime is significantly shortened, it often indicates problems such as insufficient crosslinking degree or uneven dispersion of additives. Mapping these abnormal points to the previously recorded coordinates can yield a clear "material non-uniformity map". The reason why this method can effectively solve the problem of the difficulty in detecting uneven crosslinking or abnormal molecular orientation is that it approaches from the perspective of molecular fingerprints and can provide more direct clues to the internal evolution process that is difficult to detect visually.

[0066] After screening out these non-uniform regions, stress field detection and dynamic response analysis need to be further carried out on them, including macroscopic stress distribution measurement by polarized photoelasticity and micro-region mechanical testing and environmental cyclic loading testing for stress concentration points. The intention of doing this is that even if there are subtle abnormalities in the molecular structure of certain regions, without combining with the mechanical response in the real environment, it may be impossible to determine whether it will have a significant impact on the optical performance or service life of the device. By applying simulated loads and temperature-humidity cycles at different angles to the high-stress regions, the fatigue characteristics of the material in actual use can be revealed in advance, and then identify which local defects are most likely to evolve into macroscopic cracks or delaminations. The simulated environment and multi-directional stress analysis methods used in this stage directly address the potential failure risks brought about by multi-layer bonding and micro-nano structure composites. Since resin wafers in AR / VR devices often undergo repeated mechanical deformations and temperature shocks, simple static detection cannot reflect the hidden cracks formed in this dynamic process.

[0067] Subsequently, more detailed microstructure and morphology analyses will be performed on the defect points that have been determined to be high-risk. Ultrasonic imaging can be used to determine whether there are signs of voids or delamination in a non-destructive state. If there are indeed abnormal echoes, a focused ion beam is then used to prepare a nanoscale thin slice of the target area, and information such as the local interface morphology and refractive index gradient is observed by combining scanning electron microscopy, transmission electron microscopy, or near-field optical means. The reason for implementing multiple detection methods here is that phenomena such as microcracks, interface incompatibility, or local modifier aggregation often exhibit different characteristics at different levels, such as in electron microscope images, ultrasonic reflection waves, and near-field light field distributions. Integrating this information can reproduce the true state inside the material at a very small scale. Only through such high-resolution multi-channel observations can one have an intuitive and complete understanding of the coupling effect of defects in terms of optics and mechanics. Without such microscopic analysis, even if some stress concentration areas are detected early, it is difficult to confirm whether the risk is caused by crosslinking network degradation, adhesive interface incompatibility, or filler agglomeration.

[0068] Finally, the data such as molecular spectra, mechanical responses, and microscopic defect morphologies obtained in all the above stages will be uniformly incorporated into a comprehensive evaluation system to score or classify each abnormal point, and at the same time generate an evaluation result for the entire resin wafer. With such integration, not only can one comprehensively identify which areas are likely to cause optical distortion in the short term, but also predict whether severe interlayer delamination or microcrack propagation will occur during long-term use. To ensure the reliability of the evaluation conclusion, this process often combines automated clustering, pattern recognition, and other algorithms for cross-validation. If a certain area shows obvious abnormalities in multiple detection dimensions, then it can be prioritized as a high-risk area. Through this multi-dimensional data integration with a wide coverage and high precision, rapid positioning and in-depth analysis of material inhomogeneity and local stress concentration are achieved in resin wafers with multi-layer bonding and complex micro-nano structures coexisting, fundamentally meeting the requirements for evaluating optical performance at higher resolutions and more realistic environmental conditions.

[0069] Therefore, this series of steps together construct a three-dimensional detection and analysis system from top to bottom, from macro to micro. The initial coordinate marking solves the problem of unclear detection positioning. The subsequent multi-dimensional spectral detection makes up for the deficiency of traditional detection methods in capturing the evolution of molecular states. The macro and micro mechanical tests verify the failure risks of these potential defects under actual working conditions. The final in-depth topography analysis and data fusion can connect the risk sources with the material microstructure, providing a reliable basis for the final quality control and failure warning. In this way, key issues such as molecular structure fluctuations, local stress concentration, and interlayer interface mismatch that are ignored in conventional appearance or simple transmittance tests can be discovered and accurately quantified in a timely manner, successfully filling the gap in the difficulty of characterizing the evolution and performance deterioration of multi-layer micro-nano resin wafer materials, thus effectively overcoming the core problem of a significant attenuation of optical performance caused by subtle material anomalies. Brief Description of the Drawings

[0070] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on the structures shown in these drawings.

[0071] Figure 1 It is a schematic diagram of an embodiment of the ARVR resin wafer process in the embodiment of the present invention.

[0072] The realization of the object, functional features, and advantages of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiments

[0073] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0074] It should be noted that if there are directional indications (such as up, down, left, right, front, back...) involved in the embodiments of the present invention, the directional indications are only used to explain the relative position relationship and movement conditions between components in a specific posture (as shown in the drawings). If the specific posture changes, the directional indications will also change accordingly.

[0075] In addition, the descriptions involving "first", "second", etc. in the present invention are for descriptive purposes only, and should not be construed as indicating or implying their relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In addition, "and / or" throughout the text includes three scenarios. Taking A and / or B as an example, it includes the technical solution of A, the technical solution of B, and the technical solution where both A and B are satisfied simultaneously. In addition, the technical solutions between various embodiments can be combined with each other, which must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0076] An embodiment of the present application provides an AR / VR resin wafer process. Figure 1 It is a flowchart of an AR / VR resin wafer process provided by an embodiment of the present application. In this embodiment, the method includes:

[0077] Please refer to Figure 1 , prepare a resin wafer to be measured and a reference specimen, perform positioning and marking on the micro-nano structure area of the resin wafer to be measured to obtain a detection sample;

[0078] In one embodiment of the present invention, the performing positioning and marking on the micro-nano structure area of the resin wafer to be measured to obtain a detection sample includes:

[0079] Perform dust removal, solvent cleaning and electrostatic treatment on the resin wafer to be measured, and perform a thermal and humid environment balancing treatment to obtain a pre-treated wafer;

[0080] Obtain the coordinate positions of the diffraction grating edge, the bonding layer interface and the center of the microlens array on the pre-treated wafer, and establish a detection path;

[0081] Determine reference points in the micro-nano structure area of the pre-treated wafer according to the detection path, and mark the reference points to obtain a detection sample.

[0082] The following specifically describes the steps involved in the above embodiments:

[0083] In the first step, place the resin wafer inside a protective cover, remove the surface dust using an ion air gun or oil-free dry gas, and then wipe it with a low-viscosity solvent (such as isopropyl alcohol or acetone) in cooperation with a clean cloth to ensure that the active contaminants are dissolved and removed. After the solvent cleaning is completed, an electrostatic eliminator or an electrostatic roller device can be used to reduce the surface charge potential to reduce the secondary adsorption of dust. Subsequently, the resin wafer can be placed in a temperature- and humidity-controlled environment for a fixed period of time to balance the stress and moisture content inside and on the surface of the wafer. Through these measures, the surface cleanliness and uniformity can be improved, and the internal material properties can tend to be stable. If the on-site conditions are relatively limited, the wafer can also be placed in a low-vacuum drying device to replace the temperature- and humidity-controlled environment, thereby obtaining a nearly constant surface state.

[0084] In the second step, select common devices such as an optical microscope, a laser confocal microscope, or a white light interferometer to scan the diffraction grating area on the wafer, and record the positions where the light intensity or diffraction efficiency changes significantly as the edge coordinates of the diffraction grating. Subsequently, a thickness measurement module (such as a coating thickness gauge) or a local interference imaging method can be used to identify the thickness distribution of the bonding layer, and the coordinates of the bonding layer interface can be obtained corresponding to the part with a large gradient in this distribution. To obtain the center position of the microlens array, the focal diffraction spot or the characteristics of the focused light field can be measured, and then the area with the highest light field energy concentration can be extracted as the coordinates of the center of the microlens array. After these coordinate information is aggregated in data processing software, a systematic detection path can be generated based on the shortest path algorithm or the graphic coverage algorithm, so as to ensure that key information points can be covered under the condition of limited travel range. If other alternative methods are needed, the rapid acquisition of the above coordinates can also be achieved through a secondary positioning system or a method of multi-point calibration in cooperation with an industrial camera.

[0085] In the third step, select the reference points that need to be focused on within the set of coordinates of the micro-nano structure that has been determined, and use a micron-scale marking device or an automatic marking device to perform local symbol engraving on the surface of the wafer. It is also possible to select to embed the information of these reference points into the laser etching pattern to form a clear detection mark. After the engraving is completed, a detection sample with traceable coordinates and visible marking symbols is formed. In this state, subsequent analysis and testing can be based on the same coordinate system, and each reference point can be quickly located through a vision or coordinate alignment system, reducing the error caused by repeated searches and improving the efficiency and accuracy of performing multiple detections on a certain area. Through this operation, the analysis and evaluation can be made more consistent, and the coordinate offset caused by switching between different devices can be reduced.

[0086] In an embodiment of the present invention, the obtaining of the coordinate positions of the diffraction grating edge, the bonding layer interface, and the center of the microlens array on the preprocessed wafer and the establishment of the detection path include:

[0087] Perform an optical intensity distribution detection on the diffraction grating of the preprocessed wafer to obtain an optical coupling efficiency gradient map. Determine the high-low efficiency demarcation point of the diffraction grating according to the optical coupling efficiency gradient map, and use the high-low efficiency demarcation point as the first coordinate group of the edge of the diffraction grating;

[0088] Perform a focus distribution detection on the microlens array of the preprocessed wafer to obtain a focus offset distribution map. Determine the focus distortion demarcation point of the microlens according to the focus offset distribution map, and use the focus distortion demarcation point as the second coordinate group of the center of the microlens array;

[0089] Perform a thickness uniformity detection on the bonding layer of the preprocessed wafer to obtain a thickness change rate distribution map. Determine the high-low stress demarcation point of the bonding layer according to the thickness change rate distribution map, and use the high-low stress demarcation point as the third coordinate group of the bonding layer interface;

[0090] Assign weights to the first coordinate group, the second coordinate group, and the third coordinate group, establish a coordinate priority, and generate a detection node sequence;

[0091] Generate a detection path according to the detection node sequence according to the minimum path algorithm and the maximum coverage principle.

[0092] The following specifically describes the steps involved in the above embodiments:

[0093] In the first step, use an optical probe or a laser diffraction measurement device to scan the diffraction grating area, record the intensity output values at the corresponding positions, and generate an optical energy distribution map on the coordinate plane through data visualization software. After comparing this distribution map with a preset power threshold or an optical coupling efficiency standard, the part where the optical energy is concentrated and the area where the energy rapidly decays can be identified, and the position where the two intersect is determined as the high-low efficiency demarcation line. Subsequently, the coordinate positions of these demarcation lines are uniformly collected to form a first data set. Through this operation, the spatial range of the diffraction grating functional area can be accurately identified, which is convenient for effectively covering the critical area between high efficiency and low efficiency in subsequent detections. Alternative methods include using a CCD camera in cooperation with multi-angle outgoing light scanning, which can also obtain light intensity change information similar to the above method.

[0094] In the second step, select a microlens test platform or a laser interferometer system to measure the offset of the output light spot of the microlens array point by point, and convert the test data into a focus distribution map. If obvious abnormal focus offset occurs at some positions, it is determined that the optical distortion is large at that place. According to the coordinates determined by this boundary, the high and low distortion critical points in the central area of the microlens array can be identified, and these critical points are summarized into a second data set. This method is convenient for identifying the most effective imaging range of the microlens and the problem areas prone to defocus. If the above system cannot be used under on-site conditions, the focus offset data can also be obtained by the traditional meridian plane scanning method.

[0095] In the third step, select a layer thickness gauge or white light interferometry to measure the thickness of the bonding area on the preprocessed wafer, and input the thickness value of each test point obtained into the data processing module to form a thickness change rate distribution map. If the thickness gradient of some coordinate points exceeds the pre-given range, it can be judged that the bonding stress level is significantly too high or too low at that place, and the coordinates of these obvious deviation boundaries are stored as a third data set. This operation can help lock the area where the thickness of the bonding layer fluctuates significantly, and provide a more reliable coordinate basis for subsequent analysis of whether there is peeling or compressive stress accumulation. If it is difficult to use the interference method for on-site measurement, a contact thickness measuring device can also be used for spot checks, but multiple repeated tests are required to ensure the accuracy of the data.

[0096] In the fourth step, integrate the three previously formed data sets, and different priority weights are assigned to the coordinate points in each set. If the critical area of the diffraction grating efficiency has a greater impact on the product performance, the weight value of the first data set can be increased during data processing; if particular attention is paid to the imaging deviation of the microlens, the weight of the second data set can be enhanced. Sort all the coordinate points after the assignment in the information management system, and generate a detection node list according to the comprehensive weight of the points and the density in the plane, so as to determine the order and distribution of key inspections. This can improve the utilization rate of detection resources while ensuring that key areas are not missed.

[0097] In the fifth step, according to the detection node list generated in the previous step, combine the shortest path algorithm and the coverage algorithm to formulate a test route. The shortest path algorithm is beneficial to reducing the round-trip movement distance of the detection head on the coordinate plane, and the coverage algorithm can include the edge areas where abnormalities may exist in the detection path, so as to ensure that the main functional areas are not missed under a limited travel. The finally formed path information can be directly called in the three-dimensional motion platform or imported into the automatic detection equipment in the form of a file for batch operation. Through this path planning method, more efficient detection results can be obtained in terms of time and space, laying a solid foundation for the overall quality inspection and performance verification.

[0098] Please continue to refer toFigure 1 Perform multi-dimensional spectral characterization on the test sample and the reference specimen, determine the material non-uniform distribution region of the test sample through comparative analysis, and obtain molecular structure characterization data;

[0099] In one embodiment of the present invention, the performing multi-dimensional spectral characterization on the test sample and the reference specimen, determining the material non-uniform distribution region of the test sample through comparative analysis, and obtaining molecular structure characterization data includes:

[0100] Perform Raman spectroscopy detection and infrared spectroscopy detection on the test sample to obtain first spectral data;

[0101] Perform fluorescence lifetime detection and optical anisotropy detection on the test sample to obtain second spectral data;

[0102] Perform Raman spectroscopy detection, infrared spectroscopy detection, fluorescence lifetime detection and optical anisotropy detection on the reference specimen to obtain control spectral data;

[0103] Compare and analyze the first spectral data and the second spectral data with the control spectral data respectively to determine the position coordinates of the material non-uniform distribution region and the molecular structure characterization data.

[0104] The following is a specific description of the steps involved in the above embodiments:

[0105] In the first step, select several measuring points in the marked test area, and use a Raman analysis device and an infrared spectroscopy analysis device to record the spectral information within the corresponding wave number ranges respectively. During Raman detection, a micro-area laser beam can be focused on the resin surface or internal microstructures, and by recording the intensity and displacement of the scattered signal, the cross-linking degree of the material, the dispersion of additives, and the chemical bonding state can be judged. During infrared detection, the Fourier transform method can be adopted to scan a specific wavelength range to identify common functional groups and the positions or intensities of their absorption peaks. Through the combination of the two, a more complete judgment basis for the resin formulation and cross-linking network distribution can be obtained, thus forming the first set of spectral data. This operation can finely identify the chemical composition and local modification level of the material, facilitating the capture of potential components or bonding anomalies.

[0106] In the second step, time-resolved fluorescence detection and polarization state testing are performed on the same batch of detection points. On a fluorescence lifetime device with a time-gating function, the molecular chain activity and energy transfer characteristics can be analyzed through the decay curve of the emitted optical signal over time. If the decay curves in certain regions are different from those of conventional resins, it can be inferred that there are abnormalities in the crosslinking degree or local aging phenomena. For the polarization state testing, the variation law of the transmitted or reflected optical signal can be recorded by rotating a polarizer or a polarizer array, so as to determine the degree of optical anisotropy and local orientation characteristics. These two types of data are summarized as the second group of spectral information. By cross-referencing with the first group of data, a more three-dimensional judgment can be made on the abnormal conditions at the molecular level and the optical effect level.

[0107] In the third step, the same tests are performed on the reference specimen using the same Raman device, infrared detection device, fluorescence lifetime measurement platform, and polarization analyzer, and the obtained results are stored as reference information. To avoid inaccurate conclusions caused by environmental differences or device zero-point drift, parameters such as the same laser power, scanning path, and integration time can be adopted to ensure good comparability of the reference signals. If there are standard resins with other known compositions and structures, they can also be used as weighted references to enhance the reference effectiveness of the results. The reference data established in this way can not only verify the stability of the detection system but also more quickly identify abnormal peak positions or peak intensities during subsequent comparisons.

[0108] In the fourth step, the two groups of spectral data obtained from the test samples are respectively matched with and differentially analyzed against the reference information. If significant deviations from the reference specimen are found in certain regions of the Raman, infrared, or fluorescence signals, it indicates that there are differences in the chemical bonding or molecular orientation of the materials in that region. By recording the coordinates of these abnormal positions in a graphics or data management software, a detailed map of the non-uniform distribution of the material can be generated, and the corresponding molecular structure characterization parameters can be further exported. For example, the attenuation of the signal intensity in certain bands, the shift of the Raman peak position, or the shortening of the fluorescence lifetime can all be classified and marked on the map for subsequent verification in the mechanical and morphological detection steps. This analysis process can provide more intuitive data support for resin formulation improvement, defect prediction, and its optical property evaluation.

[0109] In an embodiment of the present invention, the fluorescence lifetime detection and optical anisotropy detection are performed on the test sample to obtain the second spectral data, including:

[0110] Probe lights at different angles are applied to the test sample to obtain multi-angle fluorescence decay curves, and the molecular orientation angle distribution is calculated based on the fluorescence decay curves to obtain a molecular chain orientation state map;

[0111] Measure the polarization state of the detection sample at different incident angles to obtain a polarization degree distribution map, calculate the interfacial stress-induced birefringence value based on the polarization degree distribution map, and obtain the stress-optical coupling coefficient;

[0112] Scan the waveguide layer of the detection sample to obtain a local refractive index fluctuation map, and calculate the optical path difference distribution based on the refractive index fluctuation map and the stress-optical coupling coefficient to obtain phase delay data;

[0113] Measure the transmittance of the detection sample at different field of view angles to obtain an angle-dependent transmittance curve, and calculate the optical uniformity index based on the angle-dependent transmittance curve and the phase delay data;

[0114] Perform correlation analysis on the molecular chain orientation state map, the stress-optical coupling coefficient, the phase delay data, and the optical uniformity index to obtain second spectral data.

[0115] The following is a specific description of the steps involved in the above embodiments:

[0116] In the first step, set an excitation light source with a variable incident angle, scan point by point on the detection area, and record the curve of the fluorescence signal decay with time at each incident angle. Use a time-resolved instrument (such as a fluorescence spectrometer with a time gating function or an upconversion detection module) to collect the intensity change of the emitted light with time, and then perform normalization processing on multiple decay curves at different angles. By relating these decay curves to the molecular configuration parameters, a mapping relationship between the main axis of the molecular chain and the excitation direction can be established in the data processing stage, thereby obtaining a numerical distribution representing the degree of molecular orientation tilt. By visualizing this numerical distribution, a molecular chain orientation state map can be generated to identify whether there are abnormal orientation regions in the resin system.

[0117] In the second step, measure the transmission or reflection characteristics of the detection area for different polarized lights under multi-angle incidence conditions using a polarization analyzer or a rotating polarizer, and generate a polarization degree distribution map according to the light intensity ratio and phase shift information. If there are obvious polarization degree anomalies at certain coordinate positions, it indicates that there may be strong stress effects or molecular arrangement disturbances locally. After incorporating this polarization degree change into the optical model, the influence magnitude of the interfacial stress on birefringence can be calculated, and a coefficient related to stress can be obtained. This coefficient is used to quantify the coupling degree between mechanical strain and optical anomalies to determine whether there is a high failure risk in this micro-region.

[0118] In the third step, a precise optical scanning device is used to measure the surface or interior of the waveguide structure, and the phase information in the interference or transmission mode is combined to deduce the change amplitude of the local refractive index. Plotting this change amplitude as a refractive index fluctuation map in the coordinate distribution allows one to check for obvious refractive index anomalies or gradient bands. Combining the obtained fluctuation values with the coefficients in the previous step enables the calculation of the optical path difference and the generation of phase delay data based on this. If it is observed that the phase delay is significantly higher than the average level in certain regions, it indicates that the optical properties of these regions differ from the overall design, and further investigation of stress sources or material defects is required.

[0119] In the fourth step, light transmission measurements are performed on the detection area at different field-of-view angles, and the variation of the observed transmission intensity with the angle is plotted as a curve. By matching with the phase delay data, the overall optical uniformity index can be calculated according to existing algorithms or optical models. This index is used to evaluate whether the material can maintain stable light transmission performance and low distortion characteristics under multi-angle incidence and exit conditions. If the transmission value in a certain region drops sharply with the field-of-view angle, the uniformity index of this region will decrease significantly, indicating that imaging distortion or energy loss is likely to occur at this position in actual AR / VR applications.

[0120] In the fifth step, all the previously generated orientation state maps, stress-optical coupling coefficients, phase delay data, and optical uniformity indices are comprehensively processed to construct a visual multi-dimensional correlation model. By comparing this information in different dimensions, the corresponding relationships between the molecular chain orientation, stress concentration, optical distortion, and overall transmission quality can be revealed, thereby summarizing the second spectral data. This correlation analysis can clearly indicate which micro-regions have both obvious molecular orientation anomalies, accompanied by large phase changes and low-angle transmission performance, helping to evaluate the actual usability and potential failure modes of the resin wafer in the AR / VR optical system.

[0121] Please continue to refer to Figure 1 , perform stress field detection and dynamic response analysis on the material non-uniform distribution area to obtain mechanical property characterization data and high-risk defect points;

[0122] In an embodiment of the present invention, the performing stress field detection and dynamic response analysis on the material non-uniform distribution area to obtain mechanical property characterization data and high-risk defect points includes:

[0123] Perform polarized photoelastic detection on the material non-uniform distribution area to obtain a stress distribution map;

[0124] Perform local mechanical detection on the stress concentration points in the stress distribution map to obtain micro-region mechanical data;

[0125] Apply cyclic stress and temperature changes to the stress concentration points to obtain dynamic response characteristics;

[0126] Perform fatigue characteristic analysis based on the dynamic response characteristics to obtain fatigue characteristic data;

[0127] Integrate and analyze the micro-area mechanical data and the fatigue characteristic data to obtain mechanical property characterization data;

[0128] Determine the high-risk defect points among the stress concentration points according to the mechanical property characterization data.

[0129] The following specifically describes the steps involved in the above embodiments:

[0130] In the first step, in the detection area, a polarized photoelastic imaging device irradiates linearly polarized light and records the interference fringes of the outgoing or transmitted light to form a visual stress fringe pattern. The color or fringe density of this pattern corresponds to the internal stress state. Combining with an image recognition algorithm or a comparison scale, the overall stress intensity distribution information can be generated. In this way, the stress differences within the overall range can be identified at one time, and the local areas with higher stress values can be found from the image. This process has strong adaptability to resin materials with many layers and complex microstructures. This helps to more quickly locate the areas that need to be focused on subsequently.

[0131] In the second step, select the coordinate points with prominent stress values in the above stress fringe pattern, and use local mechanical measurement devices (such as nano-indenters or micro-area hardness testers) to quantitatively test the Young's modulus, hardness, and elastic-plastic characteristics of the material one by one. By means of point-by-point indentation or scratching, obtain the micro-area mechanical data and convert it into a unified data format, and record the actual load-bearing capacity of this area. Combining with the previously generated stress distribution map, it can be determined what kind of mechanical properties cause the stress to concentrate at this place. In this way, a more targeted judgment on the mechanical defects of the local material can be formed.

[0132] In the third step, establish a simulation environment of multi-directional load and temperature alternation at the above stress concentration points, apply repeated stress and periodic temperature rise and fall processes, and record the changes in the deformation amount or loss factor in real time. In this way, the dynamic behavior of the material under actual service conditions can be reflected within a short period, and the dynamic response information of each stress concentration point can be obtained. This method can be applied to common vibration working conditions or temperature shock environments, providing reliable parameters for identifying the potential failure risks of materials under complex conditions.

[0133] In the fourth step, substitute the deformation data or failure cycle information recorded in the previous stage into the fatigue analysis model to generate indicators such as fatigue life and energy loss to form fatigue characteristic data. If the crack initiation speed or mechanical degradation speed of the sample exceeds the safety threshold under specific conditions, it can be regarded as having potential risks. In this way, the fatigue limit performance of each high-stress point under cyclic load and temperature alternation can be clarified.

[0134] In the fifth step, the micro-region mechanical data obtained from local measurements such as nano-indentation are synchronously integrated with the fatigue characteristic data, and correlation analysis is carried out in the data management system. If certain coordinates exhibit both low local mechanical strength and a significant deterioration trend in fatigue analysis, it can be judged that these coordinates have a higher risk of failure. In this way, the microscopic mechanical parameters can be linked to the large-scale fatigue mode, and a reasonable evaluation basis can be constructed.

[0135] In the sixth step, based on the aforementioned comprehensive numerical values, the stress concentration areas are classified according to the safety level or risk coefficient, and the high-risk areas where cracks are most likely to form or peeling occurs are marked. If certain stress concentration points show abnormalities in multiple indicators, they are listed as parts that need to be repaired or replaced. This can provide a quantitative basis for structural improvement or finished product screening, making the production and maintenance processes more targeted and efficient.

[0136] In an embodiment of the present invention, applying cyclic stress and temperature changes to the stress concentration points to obtain dynamic response characteristics includes:

[0137] According to the ARVR wearing posture simulation, a multi-directional stress loading spectrum is obtained, and the multi-directional stress loading spectrum is applied to the stress concentration points to obtain a multi-dimensional stress response curve;

[0138] According to the actual use conditions of ARVR, a temperature-humidity combined action cycle is established, and the temperature-humidity combined action cycle is applied to the stress concentration points to obtain an environmental response curve;

[0139] Perform Fourier analysis on the multi-dimensional stress response curve to obtain a stress spectral density function, and calculate the cumulative damage rate according to the stress spectral density function;

[0140] The environmental response curve is segmented according to different functional regions to obtain a regional sensitivity coefficient, and the critical failure region is determined according to the regional sensitivity coefficient;

[0141] According to the corresponding relationship between the cumulative damage rate and the critical failure region, the dynamic response characteristics are obtained.

[0142] The following specifically describes the steps involved in the above embodiment:

[0143] In the first step, based on the collection and statistics of human head and facial movement data, a multi-directional mechanical load curve is constructed. These mechanical load information is input into a multi-axis loading system to apply alternating stress to a selected coordinate area. To achieve this process, a mechanical testing machine with multi-dimensional fixtures can be used to simulate stretching or bending in different directions such as up and down, left and right, front and back, etc., and arrange the loading sequence and duration in the order of wearing angle and posture switching in the test software. This can approximate the real wearing movement scenario in a short time, thereby recording the deformation amount or stress response curve of the material at different directions and time points, and integrating these curves into a multi-dimensional stress response data set.

[0144] In the second step, a multi-stage temperature and humidity cycle is set in the climate simulation device. The detection area is placed under a specific temperature rise and fall cycle and humidity change cycle. With the help of built-in strain measurement sensors or topography observation modules, the deformation and resistance changes of the material under different environmental conditions are recorded, and an environmental response curve is generated accordingly. To make the experimental results closer to the complex external conditions, different temperature and humidity peaks can be set in the program, and the corresponding holding time can be matched to reflect the external environmental fluctuations experienced by the ARVR system in repeated use or mobile scenarios. This can observe the performance stability of the material under extreme conditions such as dry heat and humidity.

[0145] In the third step, the multi-dimensional stress information changing with time collected can be mathematically expressed as a time-domain signal , whose value range is t ∈ [0, T], where T represents the recording duration. Next, the Fourier transform can be performed on this time-domain signal to obtain the spectral components in the angular frequency domain, and then calculate the stress spectral density function . If the definition method of energy spectral density is adopted, it can be written as:

[0146]

[0147] where s(t) represents the stress magnitude at time t, ω is the angular frequency, T is the total sampling duration, and j is the imaginary unit. Through this function, the stress energy distribution corresponding to different frequency bands can be observed, so as to identify the part that contributes the most to the load in the high-frequency or low-frequency range. Subsequently, according to the pre-set fatigue model or energy accumulation criterion, the energy information is mapped into the evaluation index of material fatigue damage to calculate the cumulative damage rate of the material within the same measurement period. This can quantitatively quantify the degree of fatigue impact suffered by the material under different frequency components, and provide a more targeted reference for judging whether cracks or failures will occur in the future.

[0148] In the fourth step, the environmental response curve generated by repeatedly cycling temperature and humidity is split according to the usage attributes of different functional regions in the wafer. For example, the waveguide layer region closely related to optical conduction is separately distinguished, and the curve fluctuations and strain rates therein are specifically statistically analyzed to obtain the regional sensitivity coefficient. If a certain functional region exhibits a higher performance degradation rate under the same environmental changes, a higher coefficient will be assigned to this functional region to indicate its characteristic of being prone to failure. Subsequently, the failure critical points can be screened in the coordinate plane or three-dimensional model based on these coefficients, and their precise positions can be marked in the dataset.

[0149] In the fifth step, the results of the cumulative damage rate are compared with the previously determined critical failure region. If the cumulative damage rate of some coordinate points exceeds the safety threshold and at the same time belongs to the region most sensitive to the environment, it can be inferred that these coordinate points are closer to the failure threshold. Through this process, dynamic response eigenvalues can be generated from the perspective of multi-dimensional load and environmental coupling, which are used to comprehensively evaluate the durability of the detection region under the interactive influence of multi-directional stress and temperature and humidity. Finally, these eigenvalues can be further associated with other mechanical or optical indicators in the data management system to support the quantitative judgment of the overall material performance and usage risk.

[0150] Please continue to refer to Figure 1 , and perform microstructure and morphology analysis on the high-risk defect points to obtain defect characteristic data;

[0151] In an embodiment of the present invention, the performing microstructure and morphology analysis on the high-risk defect points to obtain defect characteristic data includes:

[0152] Perform ultrasonic detection on the high-risk defect points to analyze whether there are voids, delaminations or cracks, and obtain ultrasonic imaging signals;

[0153] According to the abnormal echo positions in the ultrasonic imaging signals, precisely cut the high-risk defect points to obtain nanoscale thin film samples;

[0154] Perform interface morphology analysis on the nanoscale thin film samples to obtain material interface structure data;

[0155] Perform near-field optical detection on the high-risk defect points to obtain refractive index gradient and birefringence distribution data;

[0156] According to the material interface structure data, the refractive index gradient and birefringence distribution data, obtain defect characteristic data.

[0157] The following specifically describes the steps involved in the above embodiments:

[0158] In the first step, an acoustic scan is performed on the marked defect location using a scanning acoustic microscopy device or an immersion ultrasonic transducer, and the amplitude and arrival time of the echo signal are monitored. If there are voids, delaminations, or cracks inside the material, light and dark boundaries or abnormal points usually form in the scanned image. By converting the echo intensity and delay information into two-dimensional or three-dimensional imaging data, possible delamination interfaces or microcrack orientations can be identified around the defect points, and ultrasonic imaging information can be generated based on this.

[0159] In the second step, a focused ion beam system or a high-precision microtome device can be selected to perform micro-scale cutting on the suspected defect area in the detected sample to obtain a nano-scale thin section. If there are concerns about affecting the function of the original component, a batch of test samples with the same batch process as the finished product can be set up separately during mass production, or small-sized test blocks can be intercepted in areas with redundant space to minimize the impact on the overall structure. By means of layer-by-layer etching or precise scanning, the thickness of the thin slice can be controlled within the range of dozens of nanometers to hundreds of nanometers, so as to obtain clearer interface information in the subsequent observation stage.

[0160] In the third step, the cross-section and interlayer morphology of the prepared nano-thin slice sample are observed and recorded. A scanning electron microscope or a transmission electron microscope can be used to view the bonding characteristics between the interfaces and the microcrack orientation, and an imaging software is used to statistically analyze the flatness, void distribution, or other local structural abnormalities between the layers. This can determine whether there are potential problems such as internal structure damage or uneven adhesion, and form reliable data for quantitatively judging the interface integrity.

[0161] In the fourth step, a near-field scanning optical detection method is used at the original defect location or adjacent area. The probe is placed within a small gap range to observe the optical field distribution on the surface to obtain the local refractive index gradient and birefringence changes. If there are depressions or interface discontinuities in the physical morphology of the defect point, there will probably be obvious light bending or intensity attenuation in the near-field signal. Recording these changes and plotting the corresponding two-dimensional or three-dimensional visualization images can further illustrate the degree of influence of the defect on the overall optical performance of the material.

[0162] In the fifth step, the interface morphology data obtained in the previous stage is uniformly managed and compared with the refractive index gradient and birefringence distribution information, and indicators reflecting the nature and severity of the defects are extracted, such as crack size, interlayer void distribution, light transmission distortion degree, etc. By grouping or numericalizing the above indicators, the overall defect characteristic information can be obtained for subsequent evaluation of the degree of influence of this part on the optical imaging quality or the mechanical properties of the material.

[0163] In an embodiment of the present invention, the near-field optical detection of the high-risk defect points to obtain the refractive index gradient and birefringence distribution data includes:

[0164] The high-risk defect points are scanned and imaged at different field angles to obtain the variation curve of the point spread function with the angle, and the imaging quality attenuation coefficient is calculated according to the variation curve.

[0165] The optical field distribution of the waveguide structure around the high-risk defect points is measured to obtain the near-field light intensity distribution map, and the optical conduction loss characteristics are analyzed according to the light intensity distribution map.

[0166] The defect influence radius is established according to the imaging quality attenuation coefficient, and the refractive index is finely scanned within the influence radius to obtain a high-resolution refractive index gradient map.

[0167] The wavelet transform analysis is performed on the high-resolution refractive index gradient map to extract the local characteristic frequency, and the optical anomaly index is calculated according to the characteristic frequency.

[0168] The optical conduction loss characteristics and the optical anomaly index are comprehensively evaluated to obtain the refractive index gradient and birefringence distribution data.

[0169] The following specifically describes the steps involved in the above embodiments:

[0170] In the first step, the coordinate positions marked with high-risk problems are measured on a multi-view imaging platform. By changing different incident or observation angles, multiple sets of point spread function data are collected. To obtain the fine degree of the curve change with the angle, a laser confocal system or a high-resolution imaging device with an angle adjustment stage can be used to sequentially record the imaging spot morphology and diffusion range at each angle. After storing these data as a visualization curve, the imaging quality attenuation coefficient at different angles is calculated by means of difference or function fitting. This can intuitively reflect the influence of the defect points on the optical imaging sharpness in a large field environment.

[0171] In the second step, a coupling light source and detector combination is set at a wavelength matching the waveguide structure around the defect, and the optical field is scanned within the target coordinate range to collect the near-field light intensity distribution. To achieve this process, an optical fiber probe or a near-field scanning microscope with a micro-region positioning function can be selected, and the optical transmission and energy distribution maps are obtained through multi-point row-by-row scanning. If a significant attenuation or enhanced scattering of the light intensity is observed in a local section, it indicates that there are conduction loss characteristics at that place. Comparing this distribution with other regions can quantify the reduction degree of the waveguide light efficiency caused by the defect and provide basic data for subsequent analysis.

[0172] In the third step, using the previously obtained imaging quality attenuation values, a quantified radius value is assigned to the defect influence range. Specifically, in implementation, according to the threshold determination method, the area corresponding to when the attenuation coefficient exceeds a certain critical level can be selected as the core influence boundary. Subsequently, within this influence boundary, a higher-resolution interferometric measurement or focused light field scanning method is adopted to divide into smaller cells for refractive index measurement, so as to obtain a refined three-dimensional refractive index gradient map. This can ensure sufficient sampling density in the transition region from the defect center to the periphery, facilitating the capture of weak refractive index gradients.

[0173] In the fourth step, the refractive index distribution image obtained from the previous step is input into data processing software, and a one-dimensional or two-dimensional wavelet transform method is selected to decompose it. By setting multi-level wavelet basis functions, the local frequency components of the refractive index change at each scale are extracted, and then the characteristic values that can best represent the defect influence are screened from them and defined as characteristic frequencies. Substituting this characteristic frequency into the analysis model generates an optical anomaly index representing the degree of anomaly. The higher the value, the greater the refractive index fluctuation and distortion amplitude in this area.

[0174] In the fifth step, the previously obtained optical conduction loss characteristic curve is matched and compared with the optical anomaly index to see the coincidence or difference in coordinate and numerical distributions between the two. If a certain coordinate shows both high optical transmission attenuation and a high value in the refractive index frequency index of wavelet decomposition, then there is more reason to judge that there may be multiple mismatches or defect accumulations in this part. After exporting the detailed optical data of these coordinate points or regions, diagrams or tables of refractive index gradients and birefringence distributions can be generated, thus more precisely elaborating the impact of high-risk defects on the overall optical performance at the technical index level.

[0175] Please continue to refer to Figure 1 , and a health assessment result of the resin wafer is obtained through systematic evaluation based on the molecular structure characterization data, mechanical property characterization data, and defect characteristic data.

[0176] In an embodiment of the present invention, the systematic evaluation based on the molecular structure characterization data, mechanical property characterization data, and defect characteristic data to obtain a health assessment result of the resin wafer includes:

[0177] Performing coordinate matching on the molecular structure characterization data, mechanical property characterization data, and defect characteristic data to obtain the spatial distribution relationship of multi-source data;

[0178] Performing cross-validation on the multi-source data according to the spatial distribution relationship to determine the intersection region that shows anomalies in different detection dimensions;

[0179] Visualizing and rendering the intersection region in a three-dimensional coordinate system to generate a three-dimensional defect distribution diagram;

[0180] Calculate the local crosslinking degree deviation, stress concentration coefficient, and defect diffusion coefficient of the intersection region;

[0181] Perform a comprehensive scoring based on the local crosslinking degree deviation, stress concentration coefficient, and defect diffusion coefficient to obtain defect scoring data;

[0182] Perform a grading process on the defect scoring data to obtain the health assessment result of the resin wafer.

[0183] The following is a specific description of the steps involved in the above embodiments:

[0184] In the first step, the chemical composition or molecular bonding information, local mechanical test results, and high-resolution defect morphology data are uniformly mapped to the same coordinate system. Through an input coordinate comparison algorithm or mapping matrix, the data of each detection dimension is corresponding to a consistent position on the resin wafer. To achieve this goal, with the help of data management software or database systems, information such as Raman peak position distribution and nanoindentation point coordinates are made to correspond one by one, so as to obtain the mutual relationship between molecular characterization, mechanical characteristics, and defect morphology within the spatial range. This enables subsequent analysis to be carried out on the basis of the same spatial reference, reducing the deviation caused by different instruments or different detection methods.

[0185] In the second step, perform a multi-dimensional statistical analysis on the data that has been matched to the same coordinate system in the previous step, and screen out the coordinate ranges that show anomalies in different dimensions. If indicators such as crosslinking degree deviation, local mechanical instability, and refractive index distortion at a certain position exceed the set thresholds simultaneously, that position will be marked as a highly abnormal area in the algorithm. This process can use traditional hierarchical clustering or discriminant analysis methods, or can be automatically detected more efficiently through the integration of machine learning. This enables common abnormal points to be quickly found in the massive data and used as the subsequent key evaluation targets.

[0186] In the third step, the coordinate information and corresponding attributes of the intersection region can be imported into a three-dimensional visualization tool and presented in the three-dimensional model of the resin wafer with color or graphical features to form a three-dimensional diagram of defect distribution. By selecting an appropriate rendering method, multiple data indicators at a specific position, such as crosslinking degree or stress concentration coefficient, can be viewed on the interface, realizing the intuitive positioning of defects on different layers of the resin wafer. This can help inspection and engineering personnel clearly identify problem areas or the failure risks of specific functional areas in a visual environment.

[0187] In the fourth step, numerical calculations are performed on the molecular network anomalies, stress distribution intensities, and interface defect scales in the selected intersection regions to obtain indicators such as crosslinking degree deviation, mechanical stress aggregation, and defect diffusion rate. The specific method can be to quantify the conversion rates of certain functional groups based on infrared and Raman characteristic peaks, and combine the results of photoelasticity or micro-fatigue tests to infer the attenuation amplitude of local material elastoplasticity. Finally, the diffusion area of interface voids or cracks is statistically analyzed. This can convert abstract defect information into comparable numerical values, further improving the accuracy of the evaluation.

[0188] In the fifth step, according to the comprehensive indicators obtained in the previous step, a quantitative score is given to the defect degree of the intersection region. If the crosslinking degree deviation value or stress concentration value is relatively high, the score can be increased accordingly; if the defect diffusion value is not large, the score is appropriately reduced. This forms score data that can distinguish different defect severity levels. If the detection software has an automated analysis function, a weighted algorithm or a specific scoring model can also be used to combine the indicators into a single score, making the detection more efficient.

[0189] In the sixth step, the obtained score data is divided according to a pre-set grading interval to generate corresponding health levels for the entire resin wafer or local regions. If the score value of certain regions exceeds the warning line, they can be identified as high-risk or areas with a relatively high failure probability, and further observation or repair suggestions are given; if the score is at a low level, it can be considered relatively safe or tending to be in a qualified state. This facilitates the differentiation of products at different levels during production or inspection, and provides a direction for subsequent material and process improvement.

[0190] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structural transformation made under the inventive concept of the present invention, or direct / indirect application in other related technical fields, is included in the patent protection scope of the present invention.

Claims

1. An ARVR resin wafer process, characterized in that: include: Prepare a resin wafer to be tested and a reference test piece, and perform positioning marking on the micro-nano structure area of ​​the resin wafer to be tested to obtain a test sample; The test sample and the reference test piece are subjected to multi-dimensional spectral characterization, and the material inhomogeneous distribution area of ​​the test sample is determined by comparative analysis, and molecular structure characterization data is obtained; specifically, the method comprises: performing Raman spectroscopy detection and infrared spectroscopy detection on the test sample to obtain first spectral data; performing fluorescence lifetime detection and optical anisotropy detection on the test sample to obtain second spectral data; specifically, the method comprises: applying probe light at different angles to the test sample to obtain multi-angle fluorescence attenuation curves, calculating the molecular orientation angle distribution according to the fluorescence attenuation curves, and obtaining a molecular chain orientation state diagram; performing polarization state measurement on the test sample at different incident angles to obtain a polarization degree distribution diagram, calculating the interface stress-induced birefringence value according to the polarization degree distribution diagram, and obtaining a stress-optical coupling coefficient; scanning the waveguide layer of the test sample; Scanning to obtain a local refractive index fluctuation map, calculating the optical path difference distribution according to the refractive index fluctuation map and the stress-optical coupling coefficient, and obtaining phase delay data; measuring the transmittance of the test sample at different viewing angles to obtain an angle-dependent transmittance curve, and calculating the optical uniformity index according to the angle-dependent transmittance curve and the phase delay data; correlating and analyzing the molecular chain orientation state map, the stress-optical coupling coefficient, the phase delay data and the optical uniformity index to obtain second spectral data; performing Raman spectroscopy detection, infrared spectroscopy detection, fluorescence lifetime detection and optical anisotropy detection on the reference test piece to obtain control spectral data; comparing and analyzing the first spectral data and the second spectral data with the control spectral data, respectively, to determine the position coordinates of the material uneven distribution area and the molecular structure characterization data; The stress field detection and dynamic response analysis are performed on the unevenly distributed area of ​​the material to obtain mechanical property characterization data and high-risk defect points; specifically including: performing polarized photoelastic detection on the unevenly distributed area of ​​the material to obtain a stress distribution map; performing local mechanical detection on the stress concentration points in the stress distribution map to obtain micro-area mechanical data; applying cyclic stress and temperature changes to the stress concentration points to obtain dynamic response characteristics; specifically including: obtaining a multi-directional stress loading spectrum according to ARVR wearing posture simulation, applying the multi-directional stress loading spectrum to the stress concentration points to obtain a multi-dimensional stress response curve; establishing a temperature-humidity combined action cycle according to the actual use conditions of ARVR, applying the temperature-humidity combined action cycle to the stress concentration points The multi-dimensional stress response curve is subjected to a cooperation cycle to obtain an environmental response curve; a Fourier analysis is performed on the multi-dimensional stress response curve to obtain a stress spectrum density function, and a cumulative damage rate is calculated according to the stress spectrum density function; the environmental response curve is segmented according to different functional areas to obtain a regional sensitivity coefficient, and a critical failure area is determined according to the regional sensitivity coefficient; a dynamic response characteristic is obtained according to the corresponding relationship between the cumulative damage rate and the critical failure area; a fatigue characteristic analysis is performed according to the dynamic response characteristic to obtain fatigue characteristic data; the micro-region mechanical data and the fatigue characteristic data are integrated and analyzed to obtain mechanical characteristic characterization data; high-risk defect points in the stress concentration points are determined according to the mechanical characteristic characterization data; The high-risk defect points are subjected to microstructure and morphology analysis to obtain defect feature data; specifically including: performing ultrasonic detection on the high-risk defect points to analyze whether there are voids, peeling or cracks, and obtaining ultrasonic imaging signals; accurately cutting the high-risk defect points according to the abnormal echo positions in the ultrasonic imaging signals to obtain nanoscale thin film samples; performing interface morphology analysis on the nanoscale thin film samples to obtain material interface structure data; performing near-field optical detection on the high-risk defect points to obtain refractive index gradient and birefringence distribution data; specifically including: performing scanning imaging on the high-risk defect points at different field of view angles to obtain a curve of point spread function changing with angle, and calculating imaging quality degradation according to the curve of change reduction coefficient; measuring the light field distribution of the waveguide structure around the high-risk defect point to obtain a near-field light intensity distribution diagram, and analyzing the light conduction loss characteristics according to the light intensity distribution diagram; establishing a defect influence radius according to the imaging quality attenuation coefficient, and performing a fine refractive index scan within the influence radius to obtain a high-resolution refractive index gradient diagram; performing a wavelet transform analysis on the high-resolution refractive index gradient diagram to extract the local characteristic frequency, and calculating the optical anomaly index according to the characteristic frequency; comprehensively evaluating the light conduction loss characteristics and the optical anomaly index to obtain the refractive index gradient and birefringence distribution data; obtaining defect characteristic data according to the material interface structure data, the refractive index gradient and birefringence distribution data; A system evaluation is performed based on the molecular structure characterization data, mechanical property characterization data and defect characteristic data to obtain a health assessment result of the resin wafer.

2. The ARVR resin wafer process according to claim 1, characterized in that: The method of positioning and marking the micro-nano structure area of ​​the resin wafer to be tested to obtain a test sample includes: The resin wafer to be tested is subjected to dust removal, solvent cleaning, and electrostatic treatment, and is subjected to thermal and humidity environment balance treatment to obtain a pre-treated wafer; Obtaining the coordinate positions of the diffraction grating edge, the bonding layer interface and the center of the microlens array on the pre-processed wafer, and establishing a detection path; A reference point is determined in the micro-nano structure area of ​​the pre-processed wafer according to the detection path, and the reference point is marked to obtain a detection sample.

3. The ARVR resin wafer process according to claim 2, characterized in that: The step of obtaining the coordinate positions of the diffraction grating edge, the bonding layer interface and the center of the microlens array on the pre-processed wafer and establishing a detection path includes: Performing light intensity distribution detection on the diffraction grating of the pre-processed wafer to obtain a light coupling efficiency gradient map, determining a high- and low-efficiency dividing point of the diffraction grating according to the light coupling efficiency gradient map, and using the high- and low-efficiency dividing point as a first coordinate group of the diffraction grating edge; Performing focus distribution detection on the microlens array of the pre-processed wafer to obtain a focus offset distribution map, determining a focus distortion demarcation point of the microlens according to the focus offset distribution map, and using the focus distortion demarcation point as a second coordinate group of the center of the microlens array; Performing thickness uniformity detection on the bonding layer of the pre-processed wafer to obtain a thickness change rate distribution map, determining high and low stress demarcation points of the bonding layer according to the thickness change rate distribution map, and using the high and low stress demarcation points as the third coordinate group of the bonding layer interface; Assigning weights to the first coordinate group, the second coordinate group, and the third coordinate group, establishing coordinate priorities, and generating a detection node sequence; According to the detection node sequence, a detection path is generated according to the minimum path algorithm and the maximum coverage principle.

4. The ARVR resin wafer process according to claim 1, characterized in that: The system evaluation is performed based on the molecular structure characterization data, mechanical property characterization data and defect feature data to obtain the health evaluation result of the resin wafer, including: Coordinate matching is performed on the molecular structure characterization data, mechanical property characterization data, and defect feature data to obtain a spatial distribution relationship of multi-source data; Cross-validating the multi-source data according to the spatial distribution relationship to determine intersection areas showing abnormalities in different detection dimensions; Visually rendering the intersection area in a three-dimensional coordinate system to generate a three-dimensional defect distribution diagram; calculating a local crosslinking degree deviation, a stress concentration factor, and a defect diffusion coefficient in the intersection region; Comprehensively scoring according to the local cross-linking degree deviation, stress concentration factor and defect diffusion coefficient to obtain defect scoring data; The defect scoring data is graded to obtain a health evaluation result of the resin wafer.

Images