ARVR resin wafer technology

Abstract

The invention discloses an ARVR resin wafer process which comprises the following steps: preparing a to-be-detected resin wafer and a reference test piece, and carrying out positioning marking on a micro-nano structure area of the to-be-detected resin wafer to obtain a detection sample; performing multi-dimensional spectrum characterization on the detection sample and the reference test piece, determining a material non-uniform distribution area of the detection sample through comparative analysis, and obtaining molecular structure characterization data; carrying out 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; carrying out microstructure and morphology analysis on the high-risk defect points to obtain defect feature data; and performing system evaluation according to the molecular structure characterization data, the mechanical property characterization data and the defect characteristic data to obtain a health degree evaluation result of the resin wafer. According to the technical scheme, multi-scale characteristic characterization can be carried out on the micro-nano structure area in the resin wafer for AR / VR, and systematic detection of the characteristics such as the molecular state, the interface stress and the phase change of the material is achieved.

Description

Technical Field

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

[0002] In the development of augmented reality (AR) and virtual reality (VR) technology, resin wafers, as a key component of new optical devices, are gradually replacing traditional glass or silicon-based substrate materials due to their advantages of light weight, high plasticity, and easy micro-nano processing. This resin wafer is an indispensable core component in AR / VR optical systems, and is usually integrated with complex micro-nano structures such as diffraction gratings, waveguide layers, and microlens arrays. 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 large field of view, high transmittance, and low distortion.

[0003] However, unlike single crystals or homogeneous glass, resin wafers undergo significant molecular structural evolution during multiple curing, etching, and bonding processes: their internal cross-linked networks may form different degrees of segment orientation in different regions, and inorganic or organic modified additives may also be dispersed in an uneven manner, thereby accumulating local stress near the micro-nanostructure. More complicated, when these resin wafers are stacked or combined into multi-layer optical elements, the bonding interfaces between the layers may show both chemical incompatibility and physical stress concentration, causing microcracks or phase change regions. Such subtle changes in the material level are often not discovered in conventional appearance inspections or simple optical transmittance measurements, but will be significantly degraded in optical performance through fluctuations in refractive index or birefringence in actual use, thereby affecting the imaging quality and stability of AR / VR devices. Existing material characterization methods are mostly focused on measuring homogeneous, structure-free samples, such as conventional infrared spectroscopy, differential scanning calorimetry, stress polarization detection, etc. These methods are often unable to cope with resin wafers with complex micro-nanostructures, high-resolution waveguides or diffraction gratings. This is mainly because the etching or diffraction pattern area of ​​the micro-nano structure itself produces multiple couplings of deformation and stress. At the same time, the interface of the multi-layer composite stack is not necessarily uniform and flat in the actual functional area, showing more irregular transitions on the microscale, and conventional material testing cannot provide accurate quantification of these areas. Therefore, it is urgent to establish a complete detection method system that can simultaneously characterize the multi-scale characteristics of molecular state, interface stress, phase change, etc. in the micro-nano structure area of ​​the resin wafer for AR / VR. Summary of the invention

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

[0005] A first aspect of the present invention provides an ARVR resin wafer process, the ARVR resin wafer process comprising: 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; Performing multi-dimensional spectral characterization on the test sample and the reference test piece, determining the material uneven distribution area of ​​the test sample through comparative analysis, and obtaining molecular structure characterization data; Perform stress field detection and dynamic response analysis on the unevenly distributed area of ​​the material to obtain mechanical property characterization data and high-risk defect points; Performing microstructure and morphology analysis on the high-risk defect points to obtain defect characteristic 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.

[0006] Preferably, the positioning and marking of the micro-nano structure region of the resin wafer to be tested to obtain a test sample comprises: 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.

[0007] Preferably, 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.

[0008] Preferably, the multi-dimensional spectral characterization of the test sample and the reference test piece is performed, and the material uneven distribution area of ​​the test sample is determined by comparative analysis, and molecular structure characterization data is obtained, including: Performing Raman spectrum detection and infrared spectrum detection on the detection sample to obtain first spectrum data; Performing fluorescence lifetime detection and optical anisotropy detection on the detection sample to obtain second spectrum data; Performing Raman spectrum detection, infrared spectrum detection, fluorescence lifetime detection and optical anisotropy detection on the reference test piece to obtain reference spectrum data; The first spectral data and the second spectral data are respectively compared and analyzed with the reference spectral data to determine the position coordinates and molecular structure characterization data of the material uneven distribution area.

[0009] Preferably, the performing fluorescence lifetime detection and optical anisotropy detection on the detection sample to obtain second spectrum data includes: Applying detection light at different angles to the detection sample to obtain multi-angle fluorescence decay curves, calculating the molecular orientation angle distribution according to the fluorescence decay curves, and obtaining a molecular chain orientation state diagram; Measuring the polarization state of 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 the stress-optical coupling coefficient; Scanning the waveguide layer of the test sample 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 an optical uniformity index according to the angle-dependent transmittance curve and the phase delay data; The molecular chain orientation state diagram, stress-optical coupling coefficient, phase delay data and optical uniformity index are correlated and analyzed to obtain second spectrum data.

[0010] Preferably, the stress field detection and dynamic response analysis of the unevenly distributed area of ​​the material to obtain mechanical property characterization data and high-risk defect points includes: Performing polarized photoelastic testing on the unevenly distributed area of ​​the material to obtain a stress distribution map; Performing local mechanical testing on stress concentration points in the stress distribution diagram to obtain micro-region mechanical data; Applying cyclic stress and temperature change to the stress concentration point to obtain dynamic response characteristics; Perform fatigue characteristic analysis according to the dynamic response characteristics to obtain fatigue characteristic data; Integrate and analyze the micro-region mechanical data and the fatigue characteristic data to obtain mechanical property characterization data; The high-risk defect points among the stress concentration points are determined according to the mechanical property characterization data.

[0011] Preferably, applying cyclic stress and temperature change to the stress concentration point to obtain dynamic response characteristics includes: 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 point to obtain a multi-dimensional stress response curve; 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 point to obtain an environmental response curve; Performing Fourier analysis on the multidimensional stress response curve to obtain a stress spectrum density function, and calculating a cumulative damage rate according to the stress spectrum density function; The environmental response curve is segmented according to different functional areas to obtain regional sensitivity coefficients, and critical failure areas are determined according to the regional sensitivity coefficients; According to the corresponding relationship between the cumulative damage rate and the critical failure region, a dynamic response characteristic is obtained.

[0012] Preferably, the microstructure and morphology analysis of the high-risk defect points to obtain defect characteristic data includes: Performing ultrasonic testing on the high-risk defect points to analyze whether there are cavities, peeling or cracks, and obtaining ultrasonic imaging signals; According to the abnormal echo position in the ultrasonic imaging signal, the high-risk defect point is precisely cut to obtain a nano-scale thin slice sample; Performing interface morphology analysis on the nanoscale thin slice sample 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; Defect characteristic data are obtained according to the material interface structure data, the refractive index gradient and the birefringence distribution data.

[0013] Preferably, the near-field optical detection of the high-risk defect points to obtain refractive index gradient and birefringence distribution data includes: Scanning and imaging the high-risk defect points at different viewing angles to obtain a curve of a point spread function changing with angle, and calculating an imaging quality attenuation coefficient according to the curve; Performing optical field distribution measurement on the waveguide structure around the high-risk defect point to obtain a near-field light intensity distribution diagram, and analyzing optical 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 map; Performing wavelet transform analysis on the high-resolution refractive index gradient map, extracting local characteristic frequencies, and calculating optical anomaly indexes based on the characteristic frequencies; The light transmission loss characteristics and the optical anomaly index are comprehensively evaluated to obtain refractive index gradient and birefringence distribution data.

[0014] Preferably, 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.

[0015] The detection and analysis process of the technical solution provided in 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 area on the resin wafer. This preparation 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, various places where local stress or material inhomogeneity may be generated can be pre-positioned, making subsequent detection more targeted. If these key micro-areas are not accurately marked at the beginning, sampling bias may occur in subsequent detections, or the true material property distribution cannot be summarized because of excessive structural differences.

[0016] Subsequently, a variety of spectral methods are needed to comprehensively characterize the labeled samples and control specimens, with special attention paid to dimensions such as Raman, infrared, fluorescence lifetime, and polarization state. These methods are each sensitive to the molecular bonding mode, local cross-linking degree, fluorescence effect, and even optical anisotropy of the material. By comparing the same batch of data with the results of the control specimen, it is possible to precisely find areas where the wafer exhibits abnormalities at the molecular level. For example, if the Raman peak position of certain micro-regions is shifted, or the fluorescence lifetime is significantly shortened, it often means that there may be problems such as insufficient cross-linking and uneven dispersion of additives. By mapping these abnormal points to the previously recorded coordinates, a clear "material inhomogeneity map" can be obtained. The reason why this method can effectively solve the problem of uneven cross-linking or abnormal molecular orientation that is difficult to detect is that it cuts in from the perspective of molecular fingerprints and can provide more direct clues to internal evolution processes that are difficult to detect from the outside.

[0017] After screening out these uneven areas, they need to be further subjected to stress field detection and dynamic response analysis, including macroscopic stress distribution measurement by polarized photoelastic method, as well as micro-area mechanical testing and environmental cyclic loading testing for stress concentration points. The purpose of this is that even if subtle abnormalities appear in the molecular structure of some areas, it may not be possible to determine whether they will have a significant impact on the optical performance or service life of the device if they are not combined with the mechanical response in the real environment. By applying simulated loads and temperature and humidity cycles at different angles to high-stress areas, the fatigue characteristics of the material in actual use can be revealed in advance, and then which local defects are most likely to evolve into macroscopic cracks or peeling can be identified. The simulated environment and multi-directional stress analysis methods used in this stage directly hit the potential failure risks brought about by multi-layer bonding and micro-nanostructure composites. Because resin wafers are often subjected to repeated mechanical deformation and temperature shocks in AR / VR devices, simple static testing cannot reflect the hidden cracks formed in this dynamic process.

[0018] Later, more detailed microstructure and morphology analysis will be carried out 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 peeling in a non-destructive state. If there are abnormal echoes, a focused ion beam is used to prepare nano-scale thin slices in the target area, and local interface morphology, refractive index gradient and other information are observed by combining scanning electron microscopy, transmission electron microscopy or near-field optical means. The reason why multiple detection methods are performed here is that phenomena such as microscopic cracks, interface incompatibility or local modifier aggregation often show different characteristics at different levels such as electron microscope images, ultrasonic reflection waves, and near-field light field distribution. By integrating this information, the real state of the material inside can be reproduced on a very small scale. Only through such high-resolution multi-channel observations can we have an intuitive and complete understanding of the coupling effect of defects in optics and mechanics. Without this kind of microscopic analysis, even if certain stress concentration areas are discovered early, it is difficult to confirm whether the risks are caused by cross-linking network degradation, bonding interface incompatibility or filler agglomeration.

[0019] Finally, all the molecular spectra, mechanical responses, microscopic defect morphology and other data obtained in the above stages will be uniformly incorporated into the comprehensive evaluation system, and each abnormal point will be scored or graded, and an evaluation result for the entire resin wafer will be generated. With such integration, it is not only possible to fully identify which areas are prone to optical distortion in the short term, but also to predict whether serious interlayer delamination or microcrack expansion will occur during long-term use. In order to ensure the reliability of the evaluation conclusions, this process is often combined with automated clustering, pattern recognition and other algorithms for cross-validation. If an area has obvious abnormalities in multiple detection dimensions, it can be prioritized as a high-risk area. Through this wide coverage and high-precision multi-dimensional data integration, it is possible to quickly locate and deeply analyze material inhomogeneities and local stress concentrations in resin wafers with multi-layer bonding and complex micro-nano structures, fundamentally meeting the requirements for evaluating optical performance under higher resolution and more realistic environmental conditions.

[0020] Therefore, this series of steps together builds a three-dimensional detection and analysis system from top to bottom and from macro to micro. The initial coordinate marking solves the problem of unclear detection positioning, and the subsequent multi-dimensional spectral detection makes up for the insufficiency of traditional detection methods in capturing the evolution of molecular states. Macro and micro mechanical tests verify the failure risks of these potential defects under real working conditions. The final deep morphology analysis and data fusion can link the source of risk with the microstructure of the material, 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 in time and accurately quantified, successfully filling the gap in the difficulty of characterizing the evolution and performance degradation of multi-layer micro-nano resin wafer materials, thereby effectively overcoming the core problem of significant attenuation of optical performance due to subtle material anomalies. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0022] Figure 1 It is a schematic diagram of an embodiment of the ARVR resin wafer process in an embodiment of the present invention.

[0023] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0026] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, "and / or" in the full text includes three solutions. Taking A and / or B as an example, it includes technical solution A, technical solution B, and technical solution that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, which must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0027] An embodiment of the present application provides an ARVR resin wafer process. Figure 1 A flow chart of an ARVR resin wafer process provided in an embodiment of the present application. In this embodiment, the method includes: See also Figure 1 , preparing a resin wafer to be tested and a reference test piece, positioning and marking the micro-nano structure area of ​​the resin wafer to be tested to obtain a test sample; In one embodiment of the present invention, the step of positioning and marking the micro-nano structure region 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.

[0028] The following is a detailed description of the steps involved in the above embodiment: In the first step, the resin wafer is placed in a protective cover, and the surface dust is removed by an ion air gun or oil-free dry gas, and then wiped with a low-viscosity solvent (such as isopropyl alcohol or acetone) in combination with a clean cloth to ensure that the active dirt is dissolved and removed. After the solvent cleaning is completed, an electrostatic eliminator or electrostatic roller device can be used to reduce the surface potential to reduce the secondary adsorption of dust. Next, the resin wafer can be placed in a temperature-controlled 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 be stabilized. If the on-site conditions are more limited, the wafer can also be placed in a low vacuum drying device instead of a temperature-controlled and humidity-controlled environment to obtain a nearly constant surface state.

[0029] In the second step, common equipment such as optical microscopes, laser confocal instruments or white light interferometers are used to scan the diffraction grating area on the wafer, and the positions where the light intensity or diffraction efficiency changes significantly are recorded as the edge coordinates of the diffraction grating. Subsequently, the 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 corresponding to the portion of the distribution with a larger gradient. The center position of the microlens array can be obtained by measuring the focal diffraction spot or the focused light field characteristics, and then extracting the area with the highest light field energy concentration as the coordinates of the center of the microlens array. After these coordinate information are collected in the data processing software, a systematic detection path can be generated based on the shortest path algorithm or the graphic coverage algorithm to ensure that key information points can be covered when the travel range is limited. If other alternatives are needed, the above coordinates can also be quickly acquired through a secondary positioning system or an industrial camera combined with a multi-point calibration method.

[0030] In the third step, select the reference points that need to be focused on within the determined micro-nanostructure coordinate set, and use micron-level marking equipment or automatic marking devices to locally engrave symbols on the wafer surface. You can also choose to embed these reference point information into the laser etching pattern to form a clear detection mark. After the marking is completed, a test 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 vision or coordinate alignment systems, reducing errors caused by repeated searches and improving the efficiency and accuracy of multiple tests on a certain area. Through this operation, analysis and evaluation can be more consistent, and the coordinate offset caused by switching between different devices can be reduced.

[0031] In one embodiment of the present invention, 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.

[0032] The following is a detailed description of the steps involved in the above embodiment: In the first step, an optical probe or laser diffraction measurement device is used to scan the diffraction grating area, the intensity output value of the corresponding position is recorded, and a light energy distribution diagram is generated on the coordinate plane through data visualization software. After comparing the distribution diagram with the pre-set power threshold or light coupling efficiency standard, the part where the light energy is more concentrated and the area where the energy decays rapidly can be identified, and the position where the two meet is determined as the dividing line between high and low efficiency. Subsequently, the coordinates of the position of these dividing lines are unified to form the 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 areas of high efficiency and low efficiency in subsequent detection. Alternative methods include using a CCD camera in conjunction with multi-angle outgoing light scanning to obtain light intensity change information similar to the above method.

[0033] In the second step, a microlens test platform or a laser interferometer system is selected to measure the offset of the microlens array output light spot point by point, and the test data is converted into a focus distribution diagram. If there is an obvious focus offset anomaly at certain positions, it is determined that the location is a boundary with large optical distortion. According to the coordinates determined by this boundary, the high and low distortion critical points of the central area of ​​the microlens array can be clearly identified, and these critical points are summarized as a second data set. This method facilitates the identification of the most effective imaging range of the microlens and the problem areas prone to defocus. If the above system cannot be used due to field conditions, the focus offset data can also be obtained through the traditional meridian scanning method.

[0034] In the third step, a layered thickness gauge or white light interferometry is used to measure the thickness of the bonding area on the pre-treated wafer, and the thickness value of each test point obtained is input into the data processing module to form a thickness change rate distribution diagram. If the thickness gradient of certain coordinate points exceeds the predetermined range, it can be judged that the bonding stress level at that location is obviously too high or too low, and these coordinates that are obviously deviated from the dividing line are stored as the 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 inspection, but repeated tests are required in multiple locations to ensure the accuracy of the data.

[0035] In the fourth step, the three previously formed data sets are integrated, 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 the microlens imaging deviation is of particular concern, the weight of the second data set can be enhanced. All coordinate points after the assignment are sorted in the information management system, and a list of detection nodes is generated based on the comprehensive weight of the points and the density in the plane, thereby establishing the order and distribution of key inspections. This can improve the utilization of detection resources while ensuring that key areas are not missed.

[0036] In the fifth step, the test route is formulated based on the list of detection nodes generated in the previous step, combined with the shortest path algorithm and the coverage algorithm. The shortest path algorithm is conducive to reducing the round-trip movement distance of the detection head on the coordinate plane, while the coverage algorithm can ensure that the edge areas where abnormalities may exist are also included in the detection path, thereby ensuring that the main functional areas are not missed under limited travel. The final path information can be directly called in the three-dimensional motion platform, and can also be 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 both time and space, laying a solid foundation for overall quality inspection and performance verification.

[0037] Please continue reading Figure 1 , performing multi-dimensional spectral characterization on the test sample and the reference test piece, determining the material uneven distribution area of ​​the test sample by comparative analysis, and obtaining molecular structure characterization data; In one embodiment of the present invention, the multi-dimensional spectral characterization of the test sample and the reference test piece is performed, and the material uneven distribution area of ​​the test sample is determined by comparative analysis, and molecular structure characterization data is obtained, including: Performing Raman spectrum detection and infrared spectrum detection on the detection sample to obtain first spectrum data; Performing fluorescence lifetime detection and optical anisotropy detection on the detection sample to obtain second spectrum data; Performing Raman spectrum detection, infrared spectrum detection, fluorescence lifetime detection and optical anisotropy detection on the reference test piece to obtain reference spectrum data; The first spectral data and the second spectral data are respectively compared and analyzed with the reference spectral data to determine the position coordinates and molecular structure characterization data of the material uneven distribution area.

[0038] The following is a detailed description of the steps involved in the above embodiment: In the first step, several measuring points are selected in the marked detection area, and the spectral information within the corresponding wavenumber range is recorded using Raman analysis equipment and infrared spectrum analysis equipment. During Raman detection, the micro-area laser beam can be focused on the surface of the resin or the internal microstructure, and the intensity and displacement of the scattered signal can be recorded to determine the degree of cross-linking, dispersion of additives and chemical bonding state of the material. During infrared detection, the Fourier transform method can be used to scan a specific band range to identify common functional groups and their absorption peak positions or peak intensity differences. Through the combination of the two, a more complete basis for judging the resin formula and the distribution of the cross-linked network can be obtained, thereby forming the first set of spectral data. This operation can finely identify the chemical composition and local modification level of the material, making it easier to capture potential components or bonding anomalies.

[0039] In the second step, time-resolved fluorescence detection and polarization state test are performed on the same batch of detection points. On a fluorescence lifetime device with a time-gated function, the molecular chain activity and energy transfer characteristics can be analyzed through the decay curve of the emitted light signal over time. If the decay curve of certain areas is detected to be different from that of conventional resins, it can be inferred that the degree of cross-linking is abnormal or local aging occurs. The polarization state test can use a rotating polarizer or polarizer array to record the change pattern of the transmitted or reflected light signal, thereby determining the degree of optical anisotropy and local orientation characteristics. These two types of data are summarized as the second set of spectral information. By cross-referencing with the first set of data, a more three-dimensional judgment can be made on abnormal conditions at the molecular level and optical effect level.

[0040] In the third step, the same test is performed on the reference specimen using the same Raman equipment, infrared detection device, fluorescence lifetime measurement platform, and polarization analyzer, and the obtained results are stored as control information. In order to avoid inaccurate conclusions due to environmental differences or equipment zero drift, the same laser power, scanning path, integration time and other parameters can be adopted to ensure that the reference signal has good comparability. If there are other standard resins with known composition and structure, they can also be used as weighted controls to enhance the reference validity of the results. The control 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 in subsequent comparisons.

[0041] In the fourth step, the two sets of spectral data obtained from the test samples are matched and analyzed with the control information. If certain areas are found to deviate significantly from the reference test piece in the Raman, infrared or fluorescence signals, it means that there are differences in the chemical bonding or molecular orientation of the material in this area. By recording the coordinates of these abnormal positions in graphics or data management software, a detailed map of the uneven distribution of materials can be generated, and the corresponding molecular structure characterization parameters can be further derived. For example, the attenuation of signal intensity in certain bands, the drift of Raman peaks or the shortening of fluorescence lifetime can all be classified and marked on the graph to facilitate verification in subsequent mechanical and morphological detection links. This analysis process can provide more intuitive data support for resin formulation improvement, defect prediction and optical performance evaluation.

[0042] In one embodiment of the present invention, the step of performing fluorescence lifetime detection and optical anisotropy detection on the detection sample to obtain second spectral data includes: Applying detection light at different angles to the detection sample to obtain multi-angle fluorescence decay curves, calculating the molecular orientation angle distribution according to the fluorescence decay curves, and obtaining a molecular chain orientation state diagram; Measuring the polarization state of 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 the stress-optical coupling coefficient; Scanning the waveguide layer of the test sample 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 an optical uniformity index according to the angle-dependent transmittance curve and the phase delay data; The molecular chain orientation state diagram, stress-optical coupling coefficient, phase delay data and optical uniformity index are correlated and analyzed to obtain second spectrum data.

[0043] The following is a detailed description of the steps involved in the above embodiment: In the first step, an excitation light source with a variable incident angle is set, and the detection area is scanned point by point to record the curve of the fluorescence signal decaying over time at each incident angle. A time-resolved instrument (such as a fluorescence spectrometer with a time-gating function or an upconversion detection module) is used to collect the intensity change of the emitted light over time, and then normalize multiple decay curves at different angles. By linking these decay curves with 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 set of numerical distributions representing the degree of molecular orientation tilt. By visualizing this numerical distribution, a molecular chain orientation state diagram can be generated to identify whether abnormal orientation areas appear in the resin system.

[0044] In the second step, a polarization analyzer or a rotating polarizer is used to measure the transmission or reflection characteristics of the detection area for light of different polarizations under multi-angle incidence conditions, and a polarization distribution diagram is generated according to the light intensity ratio and phase shift information. If there is an obvious polarization anomaly at certain coordinate positions, it means that there may be strong stress or molecular arrangement disturbance in the local area. After the polarization change is included in the optical modeling, the influence of the interface stress on the birefringence can be calculated, and the coefficient related to the stress can be obtained. This coefficient is used to quantify the degree of coupling between mechanical strain and optical anomaly to determine whether the micro-area has a high risk of failure.

[0045] In the third step, a precision 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. The change amplitude is plotted as a refractive index fluctuation diagram in the coordinate distribution to check whether there are obvious refractive index anomalies or gradient bands. Combining the obtained fluctuation values ​​with the coefficients in the previous step, the optical path difference can be calculated, and phase delay data can be generated accordingly. If the phase delay is observed to be significantly higher than the average level in some areas, it means that the optical properties of the area are different from the overall design, and further investigation of stress sources or material defects is required.

[0046] In the fourth step, light transmission measurement is performed on the detection area at different field of view angles, and the observed transmission intensity is plotted as a curve as it changes with angle. By matching with the phase delay data, the overall optical uniformity index can be calculated based on the existing algorithm or optical model. This index is used to judge whether the material can maintain stable light transmission performance and low distortion characteristics under multi-angle incidence and emission conditions. If the transmission value of a certain area drops sharply with the field of view angle, the uniformity index of the area will be significantly reduced, indicating that the location is prone to imaging distortion or energy loss in actual AR / VR applications.

[0047] In the fifth step, all previously generated orientation state diagrams, stress-optical coupling coefficients, phase delay data, and optical uniformity indexes are processed comprehensively to construct a visual multidimensional correlation model. By comparing these different dimensions of information, the correspondence 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 point out which micro-regions have both obvious molecular orientation anomalies and large phase changes and low-angle transmission performance, helping to evaluate the actual usability and potential failure modes of resin wafers in AR / VR optical systems.

[0048] Please continue reading Figure 1 , performing stress field detection and dynamic response analysis on the unevenly distributed area of ​​the material to obtain mechanical property characterization data and high-risk defect points; In one embodiment of the present invention, the stress field detection and dynamic response analysis of the unevenly distributed area of ​​the material to obtain mechanical property characterization data and high-risk defect points includes: Performing polarized photoelastic testing on the unevenly distributed area of ​​the material to obtain a stress distribution map; Performing local mechanical testing on stress concentration points in the stress distribution diagram to obtain micro-region mechanical data; Applying cyclic stress and temperature change to the stress concentration point to obtain dynamic response characteristics; Perform fatigue characteristic analysis according to the dynamic response characteristics to obtain fatigue characteristic data; Integrate and analyze the micro-region mechanical data and the fatigue characteristic data to obtain mechanical property characterization data; The high-risk defect points among the stress concentration points are determined according to the mechanical property characterization data.

[0049] The following is a detailed description of the steps involved in the above embodiment: In the first step, a polarized light elastic imaging device is used to irradiate linearly polarized light in the detection area and record the emitted or transmitted interference patterns to form a visual stress fringe map. The color or fringe density of the map corresponds to the internal stress state, and the overall stress intensity distribution information can be generated in conjunction with an image recognition algorithm or a comparison scale. In this way, the stress difference within the overall range can be identified at one time, and local areas with higher stress values ​​can be found from the image. This process has a strong adaptability to resin materials with multiple layers and complex microstructures. This helps to locate areas that need to be focused on later more quickly.

[0050] In the second step, coordinate points with prominent stress values ​​are selected in the above stress fringe diagram, and the Young's modulus, hardness, and elastic-plastic characteristics of the material are quantitatively tested one by one using local mechanical measurement equipment (such as a nanoindenter or a micro-area hardness tester). By pressing or scratching point by point, the micro-area mechanical data is obtained and converted into a unified data format to record the actual bearing capacity of the area. Combined with the previously generated stress distribution map, it can be determined what kind of mechanical properties cause the stress to concentrate at this location. This can form a more targeted judgment on the mechanical defects of local materials.

[0051] In the third step, a simulation environment of multi-directional load and temperature alternation is established at the above stress concentration point, repeated stress and periodic temperature rise and fall processes are applied, and the changes in deformation or loss factor are recorded in real time. In this way, the dynamic behavior of the material under actual service conditions can be reflected in a relatively short period of time, and the dynamic response information of each stress concentration point can be obtained. This method can be applied to common vibration conditions or temperature shock environments, and provide reliable parameters for identifying the potential failure risk of materials under complex conditions.

[0052] In the fourth step, the deformation data or failure cycle information recorded in the previous stage is substituted into the fatigue analysis model to generate indicators such as fatigue life and energy loss to form fatigue characteristic data. If the crack initiation rate or mechanical degradation rate of the sample exceeds the safety threshold under specific conditions, it can be regarded as a potential risk. In this way, the fatigue limit performance of each high stress point under cyclic load and temperature alternation can be clearly determined.

[0053] In the fifth step, the micro-area mechanical data obtained by local measurements such as nanoindentation are integrated with the fatigue characteristic data, and correlation analysis is performed in the data management system. If certain coordinates show both low local mechanical strength and significant degradation trends in fatigue analysis, it can be judged that the coordinates have a higher risk of failure. In this way, the micro-mechanical parameters can be linked to the large-scale fatigue mode to build a reasonable basis for judgment.

[0054] In the sixth step, based on the above comprehensive values, the stress concentration areas are classified according to safety levels or risk factors, and the high-risk areas that are most likely to form cracks or peeling are marked. If some stress concentration points are abnormal 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 process more targeted and efficient.

[0055] In one embodiment of the present invention, applying cyclic stress and temperature change to the stress concentration point to obtain dynamic response characteristics includes: 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 point to obtain a multi-dimensional stress response curve; 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 point to obtain an environmental response curve; Performing Fourier analysis on the multidimensional stress response curve to obtain a stress spectrum density function, and calculating a cumulative damage rate according to the stress spectrum density function; The environmental response curve is segmented according to different functional areas to obtain regional sensitivity coefficients, and critical failure areas are determined according to the regional sensitivity coefficients; According to the corresponding relationship between the cumulative damage rate and the critical failure region, a dynamic response characteristic is obtained.

[0056] The following is a detailed description of the steps involved in the above embodiment: 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, and these mechanical load information is input into the multi-axis loading system to apply alternating stress to the selected coordinate area. To achieve this process, a mechanical testing machine with a multi-dimensional fixture can be used to simulate stretching or bending in different directions such as up and down, left and right, front and back, and the loading sequence and duration can be arranged in the test software according to the order of wearing angle and posture switching. In this way, the real wearable movement scenario can be approached in a short time, thereby recording the deformation or stress response curve of the material in different directions and time points, and integrating these curves into a multi-dimensional stress response data set.

[0057] In the second step, a multi-stage temperature and humidity cycle is set up in the climate simulation device, and the detection area is placed under a specific temperature rise and fall cycle and humidity change cycle. The deformation and resistance changes of the material under different environmental conditions are recorded with the help of the built-in strain measurement sensor or morphology observation module, and the environmental response curve is generated. In order 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 maintenance time can be matched to reflect the external environmental fluctuations experienced by the ARVR system in repeated use or mobile scenarios. In this way, the performance stability of the material under extreme conditions such as dry heat and humidity can be observed.

[0058] In the third step, the collected multi-dimensional stress variation information can be mathematically expressed as a time domain signal , whose value interval is t∈[0,T], where T represents the recording duration. Next, the time domain signal can be Fourier transformed to obtain the angular frequency The spectral components of the domain are then used to calculate the stress spectral density function If the definition of energy spectrum density is adopted, it can be written as: Among them, s(t) represents the stress value at time t, ω is the angular frequency, T is the total sampling time, 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. Then, according to the preset fatigue model or energy accumulation criterion, The energy information in the material is mapped to the material fatigue damage evaluation index to calculate the cumulative damage rate of the material in the same measurement period. This can quantitatively quantify the degree of fatigue impact suffered by the material under different frequency components, providing a more targeted reference for judging whether it will crack or fail in the future.

[0059] In the fourth step, the environmental response curve generated by repeated cycles of temperature and humidity is split according to the usage attributes of different functional areas in the wafer. For example, the waveguide layer area closely related to optical conduction is distinguished separately, and the curve fluctuation and strain rate are targeted for statistics to obtain the regional sensitivity coefficient. If a functional area shows a higher performance degradation rate under the same environmental changes, the functional area will be assigned a higher coefficient to indicate its prone to failure characteristics. Subsequently, the critical failure points can be screened in the coordinate plane or three-dimensional model based on these coefficients, and their exact locations can be marked in the data set.

[0060] In the fifth step, the cumulative damage rate results are compared with the previously determined critical failure areas. If the cumulative damage rate of certain coordinate points exceeds the safety threshold and belongs to the area most sensitive to the environment, it can be inferred that the coordinate point is closer to the failure threshold. Through this process, dynamic response characteristic values ​​can be generated from the perspective of multi-dimensional load and environmental coupling, which are used to comprehensively evaluate the durability of the detection area under the interaction of multi-directional stress and temperature and humidity. Ultimately, these characteristic values ​​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 risks.

[0061] Please continue reading Figure 1 , performing microstructure and morphology analysis on the high-risk defect points to obtain defect characteristic data; In one embodiment of the present invention, the microstructure and morphology analysis of the high-risk defect points to obtain defect characteristic data includes: Performing ultrasonic testing on the high-risk defect points to analyze whether there are cavities, peeling or cracks, and obtaining ultrasonic imaging signals; According to the abnormal echo position in the ultrasonic imaging signal, the high-risk defect point is precisely cut to obtain a nano-scale thin slice sample; Performing interface morphology analysis on the nanoscale thin slice sample 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; Defect characteristic data are obtained according to the material interface structure data, the refractive index gradient and the birefringence distribution data.

[0062] The following is a detailed description of the steps involved in the above embodiment: In the first step, a scanning acoustic microscopy device or an immersion ultrasonic transducer is used to acoustically scan the marked defect location and monitor the amplitude and arrival time of the echo signal. If there are cavities, delaminations or cracks inside the material, light and dark boundaries or abnormal points will usually be formed in the scanned image. By converting the echo intensity and delay information into two-dimensional or three-dimensional imaging data, possible peeling interfaces or microcrack directions can be identified around the defect point, and ultrasonic imaging information can be generated.

[0063] In the second step, a focused ion beam system or a high-precision micro-slicing device can be used to perform micro-scale cutting of suspected defective areas in the sample being tested to obtain nano-scale thin slice sections. If there is a concern about the impact on the original function, a separate batch of test samples with the same batch process as the finished product can be set up during mass production, or tiny test blocks can be cut in areas with redundant space to minimize the impact on the overall structure. By etching layer by layer or precision scanning, the thickness of the slice can be controlled within the range of tens to hundreds of nanometers, so that clearer interface information can be obtained in the subsequent observation stage.

[0064] In the third step, the cross-section and interlayer morphology of the prepared nanosheet samples are observed and recorded. Scanning electron microscopy or transmission electron microscopy can be used to view the bonding characteristics and microscopic crack directions between interfaces, and imaging software can be used to statistically analyze the flatness, gap distribution or other local structural anomalies between layers. This can determine whether there are potential problems such as damage or uneven adhesion in the internal structure, and form reliable data for quantitative judgment of interface integrity.

[0065] In the fourth step, a near-field scanning optical detection method is used at the original defect location or adjacent area, and the probe is placed in a small gap range to observe the optical field distribution of the surface to obtain the local refractive index gradient and birefringence changes. If the defect point has a depression or interface discontinuity in the physical morphology, there is a high probability that obvious light bending or intensity attenuation will appear in the near-field signal. Recording these changes and drawing corresponding two-dimensional or three-dimensional visualization images can further illustrate the extent of the defect's impact on the overall optical properties of the material.

[0066] In the fifth step, the interface morphology data obtained in the previous stage are uniformly managed and compared with the refractive index gradient and birefringence distribution information, and indicators reflecting the nature and severity of defects are extracted, such as crack size, interlayer gap distribution, light transmission distortion, etc. By grouping or digitizing the above indicators, the overall defect feature information can be obtained, which is used for subsequent evaluation of the impact of this part on the optical imaging quality or material mechanical properties.

[0067] In one embodiment of the present invention, performing near-field optical detection on the high-risk defect point to obtain refractive index gradient and birefringence distribution data includes: Scanning and imaging the high-risk defect points at different viewing angles to obtain a curve of a point spread function changing with angle, and calculating an imaging quality attenuation coefficient according to the curve; Performing optical field distribution measurement on the waveguide structure around the high-risk defect point to obtain a near-field light intensity distribution diagram, and analyzing optical 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 map; Performing wavelet transform analysis on the high-resolution refractive index gradient map, extracting local characteristic frequencies, and calculating optical anomaly indexes based on the characteristic frequencies; The light transmission loss characteristics and the optical anomaly index are comprehensively evaluated to obtain refractive index gradient and birefringence distribution data.

[0068] The following is a detailed description of the steps involved in the above embodiment: In the first step, the coordinate positions marked with high-risk problems are measured on a multi-view imaging platform, and multiple sets of point spread function data are collected by changing different incident or observation angles. In order to obtain the degree of precision of the curve as the angle changes, a laser confocal system or high-resolution imaging equipment with an angle adjustment table can be used to record the imaging spot morphology and diffusion range at each angle in turn. After storing these data as visual curves, the imaging quality attenuation coefficient at different angles is calculated by difference or function fitting. This can intuitively reflect the impact of defects on the sharpness of optical imaging in a large field of view environment.

[0069] In the second step, a coupled light source and detector combination is set at a wavelength that matches the waveguide structure around the defect, and the light field is scanned within the target coordinate range to collect the near-field light intensity distribution. To achieve this process, a fiber optic probe or a near-field scanning microscope with a micro-area positioning function can be selected to obtain light transmission and energy distribution diagrams through multi-point line-by-line scanning. If a significant attenuation of light intensity or enhanced scattering is observed in a local section, it indicates that there are conduction loss characteristics at that location. By comparing this distribution with other areas, the degree of reduction in waveguide light efficiency caused by defects can be quantified, and basic data can be provided for subsequent analysis.

[0070] In the third step, the previously obtained imaging quality attenuation value is used to assign a quantitative radius value to the defect impact range. In specific implementation, the area corresponding to when the attenuation coefficient exceeds a certain critical level can be selected as the core impact boundary according to the threshold judgment method. Subsequently, a higher-resolution interferometric measurement or focused light field scanning method is adopted within the impact boundary to divide smaller cells for refractive index measurement to obtain a refined three-dimensional refractive index gradient map. This ensures that there is sufficient sampling density in the transition area from the defect center to the periphery, which is convenient for capturing weak refractive index gradients.

[0071] In the fourth step, the refractive index distribution image obtained in the previous step is input into the data processing software, and a one-dimensional or two-dimensional wavelet transform method is selected to decompose it. By setting a multi-level wavelet basis function, the local frequency components of the refractive index changes at each scale are extracted, and then the characteristic value that best represents the influence of the defect is selected and defined as the characteristic frequency. The characteristic frequency is substituted into the analysis model to generate an optical anomaly index that represents the degree of anomaly. The higher the value, the greater the refractive index fluctuation and distortion amplitude in the area.

[0072] In the fifth step, the previously obtained light transmission loss characteristic curve is matched and compared with the optical anomaly index to see the overlap or difference between the two in coordinate and numerical distribution. If a certain coordinate shows both high light transmission attenuation and a high value in the refractive index frequency index of wavelet decomposition, there is more reason to judge that there may be multiple mismatches or defect accumulation in this part. After the detailed optical data of these coordinate points or areas are exported, a diagram or table of refractive index gradient and birefringence distribution can be generated, so as to more accurately explain the impact of high-risk defects on the overall optical performance at the technical indicator level.

[0073] Please continue reading Figure 1 , a system evaluation is performed based on the molecular structure characterization data, mechanical property characterization data and defect feature data to obtain a health assessment result of the resin wafer.

[0074] In one embodiment of the present invention, 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.

[0075] The following is a detailed description of the steps involved in the above embodiment: 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. By inputting a coordinate comparison algorithm or mapping matrix, the data of each detection dimension is mapped to a consistent position on the resin wafer. To achieve this goal, the Raman peak position distribution, nanoindentation point coordinates and other information can be mapped one by one with the help of data management software or database systems, so as to obtain the relationship between molecular characterization, mechanical characteristics and defect morphology in space. This allows subsequent analysis to be carried out on the same spatial reference basis, reducing the offset caused by different instruments or different detection methods.

[0076] In the second step, the data that has been matched to the same coordinate system in the previous step is subjected to multidimensional statistical analysis to screen out coordinate ranges that show anomalies in different dimensions. If indicators such as cross-linking degree deviation, local mechanical instability, and refractive index distortion at a certain location simultaneously exceed the set threshold, the location will be marked as a highly abnormal area in the algorithm. This process can use traditional hierarchical clustering or discriminant analysis methods, or it can be integrated with machine learning for more efficient automatic detection. In this way, common abnormal points can be quickly found in massive data and used as key evaluation targets for subsequent evaluation.

[0077] In the third step, the coordinate information and corresponding attributes of the intersection area can be imported into the 3D visualization tool, and presented in the 3D model of the resin wafer as color or graphic features to form a three-dimensional diagram of the defect distribution. By selecting the appropriate rendering method, multiple data indicators at specific locations can be viewed on the interface, such as the degree of cross-linking or the stress concentration factor, to achieve intuitive positioning of defects at different levels of the resin wafer. This can help inspection and engineering personnel clearly identify the problem-stricken areas or the risk of failure in specific functional areas in a visual environment.

[0078] In the fourth step, the molecular network anomalies, stress distribution intensity and interface defect scale in the selected intersection area are numerically calculated to obtain indicators such as cross-linking degree deviation, mechanical stress accumulation and defect diffusion rate. The specific method can be to quantify the conversion rate of certain functional groups based on infrared and Raman characteristic peaks, and combine the polarized photoelastic results or micro fatigue test results to infer the local material elastic-plastic attenuation amplitude, and finally count the diffusion area of ​​interface voids or cracks. In this way, abstract defect information can be converted into comparable values, further improving the accuracy of the evaluation.

[0079] In the fifth step, the degree of defects in the intersection area is quantitatively scored based on the comprehensive indicators obtained in the previous step. If the cross-linking 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 moderately reduced. This forms scoring data that can distinguish different defect severities. 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 to make the detection more efficient.

[0080] In the sixth step, the obtained scoring data is divided according to the pre-set grading intervals to generate corresponding health grades for the entire resin wafer or local areas. If the score value of some areas exceeds the warning line, it can be identified as a high-risk or high-probability range, and then further observation or repair suggestions are given; if the score is at a low level, it can be identified as relatively safe or biased towards a qualified state. This makes it easy to distinguish between products of different grades during production or inspection, and provides direction for subsequent material and process improvements.

[0081] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the inventive concept of the present invention, or directly / indirectly applied in other related technical fields are 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; Performing multi-dimensional spectral characterization on the test sample and the reference test piece, determining the material uneven distribution area of ​​the test sample through comparative analysis, and obtaining molecular structure characterization data; Perform stress field detection and dynamic response analysis on the unevenly distributed area of ​​the material to obtain mechanical property characterization data and high-risk defect points; Performing microstructure and morphology analysis on the high-risk defect points to obtain defect characteristic 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 multi-dimensional spectral characterization of the test sample and the reference test piece is performed, and the material uneven distribution area of ​​the test sample is determined by comparative analysis, and molecular structure characterization data is obtained, including: Performing Raman spectrum detection and infrared spectrum detection on the detection sample to obtain first spectrum data; Performing fluorescence lifetime detection and optical anisotropy detection on the detection sample to obtain second spectrum data; Performing Raman spectrum detection, infrared spectrum detection, fluorescence lifetime detection and optical anisotropy detection on the reference test piece to obtain reference spectrum data; The first spectral data and the second spectral data are respectively compared and analyzed with the reference spectral data to determine the position coordinates and molecular structure characterization data of the material uneven distribution area.

5. The ARVR resin wafer process according to claim 4, characterized in that: The step of performing fluorescence lifetime detection and optical anisotropy detection on the detection sample to obtain second spectrum data includes: Applying detection light at different angles to the detection sample to obtain multi-angle fluorescence decay curves, calculating the molecular orientation angle distribution according to the fluorescence decay curves, and obtaining a molecular chain orientation state diagram; Measuring the polarization state of 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 the stress-optical coupling coefficient; Scanning the waveguide layer of the test sample 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 an optical uniformity index according to the angle-dependent transmittance curve and the phase delay data; The molecular chain orientation state diagram, stress-optical coupling coefficient, phase delay data and optical uniformity index are correlated and analyzed to obtain second spectrum data.

6. The ARVR resin wafer process according to claim 1, characterized in that: The stress field detection and dynamic response analysis of the unevenly distributed area of ​​the material are performed to obtain mechanical property characterization data and high-risk defect points, including: Performing polarized photoelastic testing on the unevenly distributed area of ​​the material to obtain a stress distribution map; Performing local mechanical testing on stress concentration points in the stress distribution diagram to obtain micro-region mechanical data; Applying cyclic stress and temperature change to the stress concentration point to obtain dynamic response characteristics; Perform fatigue characteristic analysis according to the dynamic response characteristics to obtain fatigue characteristic data; Integrate and analyze the micro-region mechanical data and the fatigue characteristic data to obtain mechanical property characterization data; The high-risk defect points among the stress concentration points are determined according to the mechanical property characterization data.

7. The ARVR resin wafer process according to claim 6, characterized in that: The step of applying cyclic stress and temperature change to the stress concentration point to obtain dynamic response characteristics comprises: 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 point to obtain a multi-dimensional stress response curve; 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 point to obtain an environmental response curve; Performing Fourier analysis on the multidimensional stress response curve to obtain a stress spectrum density function, and calculating a cumulative damage rate according to the stress spectrum density function; The environmental response curve is segmented according to different functional areas to obtain regional sensitivity coefficients, and critical failure areas are determined according to the regional sensitivity coefficients; According to the corresponding relationship between the cumulative damage rate and the critical failure region, a dynamic response characteristic is obtained.

8. The ARVR resin wafer process according to claim 1, characterized in that: The microstructure and morphology analysis of the high-risk defect points to obtain defect characteristic data includes: Performing ultrasonic testing on the high-risk defect points to analyze whether there are cavities, peeling or cracks, and obtaining ultrasonic imaging signals; According to the abnormal echo position in the ultrasonic imaging signal, the high-risk defect point is precisely cut to obtain a nano-scale thin slice sample; Performing interface morphology analysis on the nanoscale thin slice sample 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; Defect characteristic data are obtained according to the material interface structure data, the refractive index gradient and the birefringence distribution data.

9. The ARVR resin wafer process according to claim 8, characterized in that: The near-field optical detection of the high-risk defect points to obtain refractive index gradient and birefringence distribution data includes: Scanning and imaging the high-risk defect points at different viewing angles to obtain a curve of a point spread function changing with angle, and calculating an imaging quality attenuation coefficient according to the curve; Performing optical field distribution measurement on the waveguide structure around the high-risk defect point to obtain a near-field light intensity distribution diagram, and analyzing optical 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 map; Performing wavelet transform analysis on the high-resolution refractive index gradient map, extracting local characteristic frequencies, and calculating optical anomaly indexes based on the characteristic frequencies; The light transmission loss characteristics and the optical anomaly index are comprehensively evaluated to obtain refractive index gradient and birefringence distribution data.

10. 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