Electroluminescent display illuminant packaging structure optimization system
By integrating an in-situ quantum sensing module and a self-verifying laser actuation module into a closed-loop system, the packaging structure of flexible display devices is optimized, solving the problems of interfacial strain mismatch and microcracks caused by the difference in thermal expansion coefficients between packaging film layers. This achieves high-precision chemical bonding reconstruction and improves the reliability and lifespan of flexible displays.
Patent Information
- Application Number
- CN202511129786.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-13
- Publication Date
- 2026-01-06
AI Technical Summary
In flexible display systems, the difference in thermal expansion coefficients between encapsulation film layers leads to interfacial strain mismatch and microcrack defects, resulting in accelerated moisture penetration and affecting display accuracy and reliability.
By integrating an in-situ quantum sensing module to acquire atomic displacement data of the encapsulation interface, and combining it with a materials genome engineering database and a molecular dynamics model, a stress topology cloud map is generated. Using an adversarial optimization decision module and a self-verifying laser execution module, the nanolayer distribution of the buffer layer and the barrier layer is optimized, thereby achieving real-time reconstruction of the chemical bonding state.
It effectively prevents the initiation of microcracks, improves the reliability of flexible display devices under folding conditions, completely blocks moisture penetration channels, and improves the lifespan and reliability of the display.
Smart Images
Figure CN121279062A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to an optimized system for the packaging structure of electroluminescent display light emitters. Background Technology
[0002] In electroluminescent display devices, the light emitter is the core functional unit. Its organic or inorganic semiconductor materials are highly sensitive to moisture and oxygen, making them susceptible to environmental damage, which can lead to performance degradation or permanent failure. To address this issue, the encapsulation structure achieves protection through the design of a multilayer thin film system. A typical solution includes an inorganic barrier layer that provides a highly dense physical barrier to prevent the penetration of external water and oxygen, while an organic buffer layer smooths the substrate surface and absorbs mechanical or thermal stress. These layered combinations work together to create a closed environment, maintain the integrity of the light-emitting film, and thus extend the lifespan and reliability of the display.
[0003] In the optimization process of the packaging structure of electroluminescent display emitters, the key technical challenge lies in the reliability of integrating flexible display systems into computing devices. This stems from the difference in thermal expansion coefficients between the packaging film layers, which induces interfacial strain mismatch under thermal cycling stress, leading to the formation and propagation of microcracks at the interface. For example, in computing display modules, such as foldable tablet applications, temperature fluctuations during repeated folding operations generate expansion stress differences, resulting in microcrack defects at the interface between the organic buffer layer and the inorganic barrier layer. This promotes moisture penetration, accelerates the degradation of the emitter, and thus reduces display accuracy and user interaction reliability. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this invention provides an optimization system for the packaging structure of electroluminescent display emitters. This invention solves the technical problem of moisture infiltration channels caused by interfacial strain mismatch induced by the difference in thermal expansion coefficients between packaging film layers under thermal cycling conditions, resulting in microcracks at the interface in foldable computing display modules.
[0005] To solve the above-mentioned technical problems, the specific technical solution of the present invention is as follows: The electroluminescent display light emitter packaging structure optimization system provided by the present invention includes: The in-situ quantum sensing module acquires atomic displacement data at the encapsulation interface through a quantum dot strain sensor array integrated on a flexible substrate, and connects to a materials genome engineering database to extract intrinsic parameters of hybrid materials and environmental thermodynamic parameters. The cross-scale topological evolution module uses atomic displacement data to set boundary conditions for molecular dynamics models to calculate the interface chemical bond fracture energy threshold, uses intrinsic parameters to drive the phase field model to simulate the crack propagation path, and merges the fracture energy threshold and crack propagation path to generate a macroscopic stress topological cloud map, marking the strain mismatch peak region. The adversarial optimization decision module inputs the stress topology cloud map into the adversarial network. The generator outputs the fractal configuration of the buffer layer and the nano-stacked distribution scheme of the barrier layer. The discriminator uses the stress topology cloud map to verify that the probability of microcrack initiation under folding conditions is lower than the critical threshold. The self-verifying laser execution module converts the nanolayer distribution scheme into femtosecond laser direct writing parameters. When executing the parameters, it activates photonic crystal resonance detection to obtain the phase shift characteristics of interference fringes. Based on the phase shift characteristics of interference fringes, it corrects the energy distribution of the interlayer interface and reconstructs the chemical bonding state. Among them, the environmental thermodynamic parameters are input into the cross-scale topological evolution module to adjust the initial conditions of the model, the strain mismatch peak region is input into the adversarial optimization decision module to constrain the gradient distribution of the scheme, the scheme verified by the adversarial optimization decision module is input into the self-verification laser execution module, and the bond state data reconstructed by the self-verification laser execution module is input into the in-situ quantum sensing module to measure the displacement a second time.
[0006] Furthermore, in the electroluminescent display light emitter packaging structure optimization system of the present invention, the in-situ quantum sensing module is configured to: implant a quantum dot sensor grid in the microcavity structure etched on the flexible substrate to capture fluorescence peak position drift data; Fluorescence peak position drift data are input into a confocal microscope system to analyze atomic displacement amplitude and direction; the correlation function between thermal expansion coefficient and phase transition temperature is extracted by connecting to the materials genome database, and the correlation function is applied to correct measurement errors caused by temperature fluctuations.
[0007] Furthermore, in the electroluminescent display light-emitting body packaging structure optimization system of the present invention, the cross-scale topology evolution module is configured as follows: Set boundary conditions for the molecular dynamics model using atomic displacement data, and output the critical strain energy density of chemical bonds; Using the critical strain energy density as the initial condition for crack nucleation, the phase field model is driven to generate a three-dimensional crack network topology. The critical strain energy density and the topology of the three-dimensional crack network are input into a convolutional neural network and fused to generate a macroscopic stress concentration factor cloud map.
[0008] Furthermore, in the electroluminescent display light-emitting body packaging structure optimization system of the present invention, the adversarial optimization decision module is configured as follows: In the graph neural network architecture, nodes are defined as material components, and edge weights correspond to differences in thermal expansion coefficients. Input the strain mismatch peak region of the stress topology cloud map into the graph neural network; The gradient distribution is constrained by the physical loss function to satisfy Δα / α<0.05.
[0009] Furthermore, in the electroluminescent display light-emitting body packaging structure optimization system of the present invention, the adversarial optimization decision module is configured as follows: Constructing a cellular automata environment to simulate equipment folding fatigue; Perform 2 million Monte Carlo samplings in a cellular automata environment to calculate the probability of microcrack initiation; When the probability is lower than the preset value, output a robustness verification signal for the scheme.
[0010] Furthermore, in the electroluminescent display light-emitting body packaging structure optimization system of the present invention, the self-verifying laser execution module is configured to perform beam splitting operation on the femtosecond laser pulse through a spatial light modulator; An energy envelope function is generated based on the nanolayer distribution scheme output by the adversarial optimization decision module, and a plasma waveguide structure is induced on the surface of the inorganic layer.
[0011] Furthermore, in the electroluminescent display light-emitting body packaging structure optimization system of the present invention, the self-verifying laser execution module is configured to: excite photonic crystal resonance to detect the cavity mode shift of the reflection spectrum; When the offset exceeds 0.5nm, a second femtosecond laser scan is triggered; Adjust the femtosecond laser single pulse energy to the preset bond energy threshold to complete the siloxane bond reconstruction.
[0012] Furthermore, the electroluminescent display light-emitting element packaging structure optimization system of the present invention further includes: The direction of the atomic displacement vector output by the in-situ quantum sensing module is input into the molecular dynamics model of the cross-scale topological evolution module and set as the boundary condition; The measurement error data corrected by the in-situ quantum sensing module drives the cross-scale topological evolution module to update the intrinsic parameters.
[0013] Furthermore, the electroluminescent display light-emitting element packaging structure optimization system of the present invention further includes: The strain mismatch peak region output by the cross-scale topology evolution module is input into the adversarial optimization decision module generator as a gradient distribution constraint boundary. The three-dimensional crack network topology input of the cross-scale topology evolution module is used as the discriminator of the adversarial optimization decision module to drive the adjustment of the crack sensitivity coefficient.
[0014] Furthermore, the electroluminescent display light-emitting element packaging structure optimization system of the present invention further includes: The buffer layer fractal geometry generated by the adversarial optimization decision module is input into the self-verifying laser execution module; The femtosecond laser direct writing angle is set according to the fractal dimension of the fractal geometry, ranging from 1.2 to 1.8.
[0015] Beneficial effects of this invention; This invention captures atomic displacement data of the encapsulation interface and corrects temperature errors through an in-situ quantum sensing module, providing high-precision deformation input; a cross-scale topological evolution module integrates microscopic bond fracture energy and mesoscopic crack paths to generate a macroscopic stress cloud map, accurately identifying strain mismatch regions; an adversarial optimization decision module constrains gradient distribution with a physical loss function and verifies the microcrack probability threshold, generating thermally expansion-compatible fractal configurations and nanolayered schemes; a self-verifying laser execution module synchronizes photon detection and femtosecond laser correction, reconstructs the bonding interface in real time, and provides feedback to verify displacement; all modules form a closed loop of "measurement, modeling, optimization, manufacturing, and verification," making the entire process of thermal expansion mismatch controllable from atomic displacement perception to interface reconstruction, reducing the probability of microcrack initiation to near zero, completely blocking moisture penetration channels, and improving the reliability of flexible display devices under folding conditions. Attached Figure Description
[0016] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on the drawings without creative effort.
[0017] Figure 1 The system architecture diagram of the electroluminescent display light-emitting body packaging structure optimization system provided in the embodiments of the present invention is shown. Detailed Implementation
[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only one module of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The technical solutions provided by various embodiments of this invention will be described in detail below with reference to the accompanying drawings. To better understand the objectives of this invention, it will be described in further detail below.
[0019] Please see Figure 1 The electroluminescent display light emitter packaging structure optimization system provided by the present invention includes: The in-situ quantum sensing module acquires atomic displacement data at the encapsulation interface through a quantum dot strain sensor array integrated on a flexible substrate, and connects to a materials genome engineering database to extract intrinsic parameters of hybrid materials and environmental thermodynamic parameters. The cross-scale topological evolution module uses atomic displacement data to set boundary conditions for molecular dynamics models to calculate the interface chemical bond fracture energy threshold, uses intrinsic parameters to drive the phase field model to simulate the crack propagation path, and merges the fracture energy threshold and crack propagation path to generate a macroscopic stress topological cloud map, marking the strain mismatch peak region. The adversarial optimization decision module inputs the stress topology cloud map into the adversarial network. The generator outputs the fractal configuration of the buffer layer and the nano-stacked distribution scheme of the barrier layer. The discriminator uses the stress topology cloud map to verify that the probability of microcrack initiation under folding conditions is lower than the critical threshold. The self-verifying laser execution module converts the nanolayer distribution scheme into femtosecond laser direct writing parameters. When executing the parameters, it activates photonic crystal resonance detection to obtain the phase shift characteristics of interference fringes. Based on the phase shift characteristics of interference fringes, it corrects the energy distribution of the interlayer interface and reconstructs the chemical bonding state. Among them, the environmental thermodynamic parameters are input into the cross-scale topological evolution module to adjust the initial conditions of the model, the strain mismatch peak region is input into the adversarial optimization decision module to constrain the gradient distribution of the scheme, the scheme verified by the adversarial optimization decision module is input into the self-verification laser execution module, and the bond state data reconstructed by the self-verification laser execution module is input into the in-situ quantum sensing module to measure the displacement a second time.
[0020] The in-situ quantum sensing module involves etching a microcavity structure array on the surface of a flexible substrate and implanting CdSe / ZnS core-shell quantum dots to form a distributed sensor grid. When atomic displacement occurs at the encapsulation interface, the quantum dot lattice distortion induces a shift in the fluorescence spectrum peak position. The shift spectrum is captured by a confocal microscopy system, and the lattice distortion energy is calculated using density functional theory to invert the atomic displacement amplitude and vector direction. Simultaneously, the module connects to a materials genome engineering database, retrieves the entropy-stable phase diagram based on the hybrid material molecular formula, extracts the correlation function between the thermal expansion coefficient and the phase transition temperature, and applies the correlation function to correct the quantum dot fluorescence measurement error caused by environmental temperature fluctuations. The module outputs high-precision atomic displacement data and intrinsic material parameters.
[0021] The cross-scale topological evolution module sets the boundary conditions of the molecular dynamics model using the atomic displacement vector output by the in-situ quantum sensing module. It uses the ReaxFF reactive force field to simulate the chemical bond breaking process at the organic / inorganic interface and calculates the critical strain energy density at which the bond energy decays to the failure threshold. Using the critical strain energy density as the initial condition for crack nucleation, it drives the Cahn-Hilliard phase field model to generate a three-dimensional crack network topological path. By fusing the critical strain energy density tensor and crack network topological data through a convolutional neural network, the atomic-scale dislocation motion is mapped into a macroscopic stress concentration factor cloud map, which identifies the peak region of strain mismatch.
[0022] The adversarial optimization decision module inputs the strain mismatch peak region of the stress topology map into the graph neural network architecture, defining the nodes as polyimide buffer layers or... The material composition of the barrier layer and the edge weights correspond to the differences in thermal expansion coefficients; the generator forces the gradient distribution to satisfy Δα / α<0.05 through the physical loss function, and outputs the fractal geometry of the buffer layer and the periodic distribution scheme of the nano-stack of the barrier layer; the discriminator constructs a cellular automata environment to simulate folding fatigue, performs Monte Carlo sampling to calculate the probability of microcrack initiation, and outputs a robustness verification signal of the scheme when the probability is lower than the critical threshold.
[0023] The self-verifying laser execution module converts the nanolayered distribution scheme into a femtosecond laser spatial energy envelope function, and splits the laser pulses through a spatial light modulator. When inducing a plasma waveguide structure on the inorganic layer surface, it simultaneously excites a photonic crystal resonance to detect the cavity mode shift of the reflection spectrum. When the shift exceeds a set threshold, it triggers a second femtosecond laser scan and adjusts the single pulse energy to the bond energy threshold, so that the siloxane bonding interface is reconstructed to the target state. After reconstruction, the bonding state data is fed back to the in-situ quantum sensing module for secondary verification of the displacement.
[0024] Specifically, in the electroluminescent display light emitter packaging structure optimization system of the present invention, the in-situ quantum sensing module is configured to: implant a quantum dot sensor grid in the microcavity structure etched on a flexible substrate to capture fluorescence peak position drift data. Fluorescence peak position drift data are input into a confocal microscope system to analyze atomic displacement amplitude and direction; the correlation function between thermal expansion coefficient and phase transition temperature is extracted by connecting to the materials genome database, and the correlation function is applied to correct measurement errors caused by temperature fluctuations.
[0025] The in-situ quantum sensing module etches a microcavity structure array on the surface of a flexible substrate, implants CdSe / ZnS core-shell quantum dots to form a distributed sensor grid, and captures fluorescence peak position drift data induced by encapsulation interface strain. The fluorescence peak position drift data is input into a confocal microscopy system, and the mapping relationship between lattice distortion energy and displacement amplitude is calculated using density functional theory to resolve the atomic displacement vector direction. The correlation function between the thermal expansion coefficient and phase transition temperature is extracted by connecting to a materials genome engineering database. The correlation function is then applied to correct the interference error of ambient temperature fluctuations on the quantum dot fluorescence signal, and high-precision atomic displacement data is output.
[0026] Specifically, in the electroluminescent display light-emitting body packaging structure optimization system of the present invention, the cross-scale topology evolution module is configured as follows: Set boundary conditions for the molecular dynamics model using atomic displacement data, and output the critical strain energy density of chemical bonds; Using the critical strain energy density as the initial condition for crack nucleation, the phase field model is driven to generate a three-dimensional crack network topology. The critical strain energy density and the topology of the three-dimensional crack network are input into a convolutional neural network and fused to generate a macroscopic stress concentration factor cloud map.
[0027] The cross-scale topology evolution module sets the boundary conditions of the molecular dynamics model using atomic displacement data output from the in-situ quantum sensing module, simulates the chemical bond breaking process at the interface using a reactive force field, and outputs the critical strain energy density at which the bond energy decays to the failure threshold. Using the critical strain energy density as the initial condition for crack nucleation, the phase field model is driven to generate a three-dimensional crack propagation path network topology. The critical strain energy density tensor and the crack network topology data are input into a convolutional neural network and fused to generate a macroscopic stress concentration coefficient cloud map, which identifies the peak region of strain mismatch at the organic / inorganic layer interface.
[0028] Specifically, in the electroluminescent display light-emitting body packaging structure optimization system of the present invention, the adversarial optimization decision module is configured as follows: In the graph neural network architecture, nodes are defined as material components, and edge weights correspond to differences in thermal expansion coefficients. Input the strain mismatch peak region of the stress topology cloud map into the graph neural network; The gradient distribution is constrained by the physical loss function to satisfy Δα / α<0.05.
[0029] The generator of the adversarial optimization decision module defines nodes as polyimide buffer layers or in the graph neural network architecture. The material composition of the barrier layer and the edge weights correspond to the differences in the thermal expansion coefficients of the materials. The strain mismatch peak region output by the cross-scale topology evolution module is input into the graph neural network. The gradient distribution is forced to satisfy Δα / α<0.05 through the physical loss function, and the fractal geometry of the buffer layer and the periodic distribution scheme of the nano-stack of the barrier layer are output.
[0030] Specifically, in the electroluminescent display light-emitting body packaging structure optimization system of the present invention, the adversarial optimization decision module is configured as follows: Constructing a cellular automata environment to simulate equipment folding fatigue; Perform 2 million Monte Carlo samplings in a cellular automata environment to calculate the probability of microcrack initiation; When the probability is lower than the preset value, output a robustness verification signal for the scheme.
[0031] The discriminator of the adversarial optimization decision module constructs a cellular automaton environment to simulate the fatigue condition of the hinge area of the foldable device. Monte Carlo sampling is performed in the environment to calculate the probability of microcrack initiation. When the probability is lower than the critical threshold, a robustness verification signal of the gradient material scheme is output, triggering the generator to lock the final optimization scheme.
[0032] Specifically, in the electroluminescent display light emitter packaging structure optimization system of the present invention, the self-verifying laser execution module is configured to perform beam splitting operation on femtosecond laser pulses through a spatial light modulator; An energy envelope function is generated based on the nanolayer distribution scheme output by the adversarial optimization decision module, and a plasma waveguide structure is induced on the surface of the inorganic layer.
[0033] The self-verifying laser execution module performs beam splitting on femtosecond laser pulses using a spatial light modulator, generates an energy envelope function based on the nanolayered distribution scheme output by the adversarial optimization decision module, and induces a plasma waveguide structure on the surface of the inorganic barrier layer to achieve precise deposition of the nanolayered periodic structure.
[0034] Specifically, in the electroluminescent display light-emitting body packaging structure optimization system of the present invention, the self-verifying laser execution module is configured to: excite photonic crystal resonance to detect the cavity mode shift of the reflection spectrum; When the offset exceeds 0.5nm, a second femtosecond laser scan is triggered; Adjust the femtosecond laser single pulse energy to the preset bond energy threshold to complete the siloxane bond reconstruction.
[0035] The photonic crystal resonance is excited to detect the Fabry-Perot cavity mode shift in the reflection spectrum; when the shift exceeds the set threshold, a femtosecond laser secondary scanning operation is triggered; the single pulse energy is adjusted to the preset bond energy threshold to complete the reconstruction of the siloxane chemical bonding interface.
[0036] Specifically, the electroluminescent display light-emitting body packaging structure optimization system of the present invention further includes: The direction of the atomic displacement vector output by the in-situ quantum sensing module is input into the molecular dynamics model of the cross-scale topological evolution module and set as the boundary condition; The measurement error data corrected by the in-situ quantum sensing module drives the cross-scale topological evolution module to update the intrinsic parameters.
[0037] The atomic displacement vector direction output by the in-situ quantum sensing module is input into the molecular dynamics model of the cross-scale topological evolution module and set as the simulated boundary condition. The measurement error data corrected by the in-situ quantum sensing module drives the cross-scale topological evolution module to update the intrinsic parameters of the hybrid material, eliminating the influence of environmental noise on crack path prediction.
[0038] Specifically, the electroluminescent display light-emitting body packaging structure optimization system of the present invention further includes: The strain mismatch peak region output by the cross-scale topology evolution module is input into the adversarial optimization decision module generator as a gradient distribution constraint boundary. The three-dimensional crack network topology input of the cross-scale topology evolution module is used as the discriminator of the adversarial optimization decision module to drive the adjustment of the crack sensitivity coefficient.
[0039] The strain mismatch peak region output by the cross-scale topology evolution module is input into the adversarial optimization decision module generator as a gradient distribution constraint boundary; the three-dimensional crack network topology input discriminator of the cross-scale topology evolution module drives the adjustment of crack sensitivity coefficient weights, so that the optimization scheme avoids high failure probability regions.
[0040] Specifically, the electroluminescent display light-emitting body packaging structure optimization system of the present invention further includes: The buffer layer fractal geometry generated by the adversarial optimization decision module is input into the self-verifying laser execution module; The femtosecond laser direct writing angle is set according to the fractal dimension of the fractal geometry, ranging from 1.2 to 1.8.
[0041] The buffer layer fractal geometry generated by the adversarial optimization decision module is input into the self-verifying laser execution module; the femtosecond laser direct writing angle is set according to the fractal dimension range of the fractal geometry, matching the beam incident angle parameters of different dimension configurations.
[0042] This invention addresses the problems of interfacial strain mismatch and microcrack penetration caused by differences in the thermal expansion coefficients between encapsulation film layers in flexible displays by constructing a closed-loop system of "quantum sensing, cross-scale modeling, adversarial optimization, and self-verification execution." The specific technical logic is as follows: An in-situ quantum sensing module integrates a quantum dot sensor array on a flexible substrate to capture fluorescence peak shifts caused by atomic displacements at the encapsulation interface. A confocal microscopy system analyzes the shift data to invert the displacement amplitude and direction. A materials genome database provides a correlation function between the coefficient of thermal expansion and the phase transition temperature, correcting for measurement errors caused by environmental temperature fluctuations, and outputting high-precision atomic displacement vectors and intrinsic material parameters. This step overcomes the limitation of existing technologies in measuring nanoscale interface deformation.
[0043] The cross-scale topological evolution module sets molecular dynamics boundary conditions using corrected atomic displacement data to simulate the chemical bond breaking energy threshold at the organic / inorganic interface; it then uses the critical strain energy density to drive the phase field model to generate a three-dimensional crack propagation path; a convolutional neural network fuses microscopic bond breaking data with mesoscopic crack topology, outputting a macroscopic stress concentration cloud map to identify strain mismatch peak regions. This step establishes a complete causal chain of atomic displacement, bond breaking, crack propagation, and macroscopic strain.
[0044] The adversarial optimization decision-making module inputs the strain mismatch peak region into the graph neural network: the generator defines buffer / barrier layer material nodes, forces the thermal expansion gradient distribution to satisfy Δα / α<0.05 through a physical loss function, and outputs fractal configurations and nanolayered schemes; the discriminator simulates folding fatigue in a cellular automata environment, and Monte Carlo sampling verifies that the probability of microcracks is below the critical threshold. This step generates a non-obvious gradient material structure, eliminating thermal expansion mismatch.
[0045] The self-verifying laser execution module converts the optimized scheme into a femtosecond laser energy envelope, and the spatial light modulator uses beam splitting pulses to induce a plasma waveguide structure. Simultaneously, a photonic crystal resonates to detect the cavity mode offset. When the offset exceeds a threshold, a secondary laser scan is triggered, and the energy is adjusted to the bond energy threshold to reconstruct the siloxane bonding interface. The reconstructed data is then fed back to the quantum sensing module for secondary verification of the displacement. This step enables simultaneous suppression of crack formation during the manufacturing process.
[0046] Specific embodiments of the present invention are as follows: In foldable tablet applications, the difference in thermal expansion coefficients between the encapsulation film layers causes microcracks to form at the interface during repeated folding, creating moisture permeation channels. This invention is implemented according to the following steps: In-situ quantum sensing modules are fabricated by etching a microcavity array in the hinge region of a flexible substrate and embedding a CdSe / ZnS quantum dot sensor grid. During device folding operations, the quantum dots capture the fluorescence peak shifts induced by interfacial atomic displacements; confocal microscopy is used to analyze the magnitude and direction of these shifts. Simultaneously, a materials genome database is connected to extract polyimide-... The thermal expansion correlation function of the hybrid system corrects for environmental temperature fluctuation errors and outputs accurate atomic displacement vectors.
[0047] The cross-scale topology evolution module sets molecular dynamics boundary conditions using displacement vectors, simulates the chemical bond breaking process at the organic / inorganic interface in the hinge region, and outputs the critical strain energy density; it drives the phase field model to generate a three-dimensional crack network topology; the convolutional neural network fuses the critical strain energy and crack path data, and outputs a stress concentration cloud map of the hinge region to identify the maximum strain mismatch site.
[0048] The adversarial optimization decision module inputs the strain mismatch peak into the graph neural network. The generator defines the weights of the fractal nodes of the buffer layer and the stacked edges of the blocking layer. The physical loss function constrains the distribution of the thermal expansion gradient. The discriminator simulates 200,000 folding fatigue cycles in a cellular automaton and outputs the fractal configuration scheme when the crack probability is lower than the critical threshold using Monte Carlo calculation.
[0049] The self-verifying laser execution module conversion scheme uses a femtosecond laser energy envelope, with the spatial light modulator splitting pulses in... Layer-induced plasma waveguide; synchronous photonic crystal resonance detection of cavity mode offset; when the offset exceeds the threshold, secondary scanning adjusts the energy to the bond energy threshold and reconstructs the siloxane bonding interface; the quantum sensing module verifies the displacement to be close to zero in the second step.
[0050] By eliminating quantum sensing errors to improve modeling accuracy, analyzing crack physics mechanisms through cross-scale models, generating non-obvious gradient materials through adversarial optimization, and achieving atomic-level reconstruction through femtosecond laser synchronous detection, a closed loop is formed to suppress microcrack initiation and block the moisture penetration path of foldable devices.
[0051] The in-situ quantum sensing module embeds a quantum dot sensor grid within a microcavity structure etched onto a flexible substrate. It analyzes the atomic displacement vector at the encapsulation interface by capturing fluorescence peak drift data. A correlation function between the coefficient of thermal expansion and phase transition temperature, provided by a materials genome engineering database, directly corrects for measurement errors caused by environmental temperature fluctuations, ensuring the accuracy of the atomic displacement data. The corrected data is then input into a cross-scale topological evolution module, where the atomic displacement vector direction serves as a boundary condition for the molecular dynamics model. This drives the ReaxFF reactive force field to calculate the chemical bond breaking energy threshold at the organic / inorganic interface. Simultaneously, the corrected environmental thermodynamic parameters update the model's initial conditions, eliminating the influence of thermal noise on the simulation.
[0052] The cross-scale topology evolution module uses the critical strain energy density of chemical bonds as the initial condition for crack nucleation in the phase-field model, generating a three-dimensional crack network topology path. A convolutional neural network fuses the critical strain energy density tensor with crack topology data, outputting a macroscopic stress concentration coefficient contour map to identify strain mismatch peak regions. This contour map is input into the graph neural network architecture of the adversarial optimization decision module, where nodes are defined as the material composition of the polyimide buffer layer or the silica barrier layer, and edge weights correspond to differences in thermal expansion coefficients. A physical loss function forces the gradient distribution to satisfy a relative difference in thermal expansion coefficients of less than 5%, generating the fractal geometry of the buffer layer and the nanolayered distribution scheme of the barrier layer.
[0053] The discriminator in the adversarial optimization decision-making module constructs a cellular automata environment to simulate the repeated folding conditions of the hinge region of a foldable device, performing Monte Carlo sampling to calculate the probability of microcrack initiation. When the probability falls below a critical threshold, a robustness verification signal is output, triggering the generator to lock the fractal configuration and nanolayered scheme. The three-dimensional crack network topology output from the cross-scale topology evolution module is synchronously input into the discriminator, driving the adjustment of crack sensitivity coefficient weights to avoid high failure probability regions.
[0054] The self-verifying laser execution module converts the nanolayered distribution scheme into a femtosecond laser spatial energy envelope function. A spatial light modulator splits the laser pulses to induce a plasmonic waveguide structure on the surface of the silica barrier layer. Simultaneously, a photonic crystal resonance is excited to detect the cavity mode shift in the reflection spectrum. When the shift exceeds a set threshold, a second femtosecond laser scan is triggered, and the single-pulse energy is adjusted to the siloxane bond energy threshold, reconstructing the chemical bonding interface. The reconstructed bonding state data is then fed back to the in-situ quantum sensing module for a second displacement measurement, forming a closed-loop verification.
[0055] The fractal geometry of the buffer layer generated by the adversarial optimization decision-making module is input into the self-verifying laser execution module. The fractal dimension is set in the range of 1.2 to 1.8, which directly determines the femtosecond laser direct writing angle parameters and matches the absorption characteristics of the fractal structure for folding stress. The closed-loop data flow between modules achieves full-process control of thermal expansion mismatch in foldable displays from atomic level perception to interface reconstruction through atomic displacement measurement, stress cloud map generation, gradient scheme optimization, laser interface reconstruction, and secondary verification, thus blocking moisture penetration channels.
Claims
1. An electroluminescent display light emitter package structure optimization system, characterized by, The method comprises the following steps: An in-situ quantum sensing module acquires atomic displacement data of a packaging interface through a quantum dot strain sensor array integrated in a flexible substrate, and connects a material genetic engineering database to extract intrinsic parameters and environmental thermodynamic parameters of a hybrid material; A cross-scale topological evolution module sets boundary conditions of a molecular dynamics model with the atomic displacement data to calculate an interface chemical bond breaking energy threshold, drives a phase field model to simulate a crack propagation path with the intrinsic parameters, and generates a macroscopic stress topological cloud map by fusing the breaking energy threshold and the crack propagation path to identify a strain mismatch peak area; An adversarial optimization decision module inputs the stress topological cloud map into a generative adversarial network, and the generator outputs a buffer layer fractal configuration and a barrier layer nanolaminate distribution scheme, and the discriminator verifies that the scheme has a micro-crack initiation probability lower than a critical threshold under folding conditions; A self-verification laser execution module converts the nanolaminate distribution scheme into femtosecond laser direct writing parameters, activates a photonic crystal resonance detection to obtain interference fringe phase shift characteristics when the parameters are executed, corrects the interfacial energy distribution according to the interference fringe phase shift characteristics, and reconstructs the chemical bonding state. The environmental thermodynamic parameters are input into the cross-scale topological evolution module to adjust the initial conditions of the model, the strain mismatch peak area is input into the adversarial optimization decision module to constrain the gradient distribution of the scheme, the scheme verified by the adversarial optimization decision module is input into the self-verification laser execution module, and the reconstructed bonding state data of the self-verification laser execution module is input into the in-situ quantum sensing module for secondary measurement of displacement.
2. The electroluminescent display emitter package structure optimization system of claim 1, wherein, The in-situ quantum sensing module is configured to implant a quantum dot sensor grid in a flexible substrate etched microcavity structure to capture fluorescence peak shift data; The fluorescence peak shift data is input into a confocal microscopy system to analyze the atomic displacement amplitude and direction; The thermal expansion coefficient and the phase transition temperature correlation function are extracted from the material genetic database, and the correlation function is applied to correct the measurement error caused by temperature fluctuations.
3. The electroluminescent display emitter package structure optimization system of claim 2, wherein, The cross-scale topological evolution module is configured to: Set the boundary conditions of the molecular dynamics model with the atomic displacement data, and output the critical strain energy density of the chemical bond; Use the critical strain energy density as the initial condition of crack nucleation to drive the phase field model to generate a three-dimensional crack network topology; Input the critical strain energy density and the three-dimensional crack network topology into a convolutional neural network to generate a macroscopic stress concentration coefficient cloud map.
4. The electroluminescent display emitter package structure optimization system of claim 3, wherein, The adversarial optimization decision module is configured to: Define the nodes as material components and the edge weights as the difference in thermal expansion coefficient in the graph neural network architecture; Input the strain mismatch peak area of the stress topological cloud map into the graph neural network; Constrain the gradient distribution to satisfy Δα / α<0.05 through a physical loss function.
5. The electroluminescent display light emitter package structure optimization system of claim 4, wherein, The adversarial optimization decision module is configured to: Simulate device folding fatigue in a cellular automaton environment; Perform 2 million Monte Carlo samplings in the cellular automaton environment to calculate the micro-crack initiation probability; When the probability is lower than a preset value, output a scheme robustness verification signal.
6. The electroluminescent display emitter package structure optimization system of claim 5, wherein, The self-verification laser execution module is configured to perform beam splitting of a femtosecond laser pulse through a spatial light modulator; Generate an energy envelope function according to the nanolaminate distribution scheme output by the adversarial optimization decision module to induce a plasmonic waveguide structure on the surface of an inorganic layer.
7. The electroluminescent display emitter package structure optimization system of claim 6, wherein, The self-verification laser execution module is configured to excite a photonic crystal resonance detection reflection spectrum cavity to detect a cavity mode shift; When the shift exceeds 0.5 nm, a femtosecond laser is triggered for secondary scanning; The single pulse energy of the femtosecond laser is adjusted to a preset bond energy threshold, and the siloxane bonding reconstruction is completed.
8. The electroluminescent display emitter package structure optimization system of claim 7, wherein, Further comprising: The atomic displacement vector direction output by the in-situ quantum sensing module is input into the molecular dynamics model of the cross-scale topological evolution module, which is set as a boundary condition; The measurement error data corrected by the in-situ quantum sensing module drives the cross-scale topological evolution module to update intrinsic parameters.
9. The electroluminescent display emitter package structure optimization system of claim 8, wherein, Further comprising: The strain mismatch peak region output by the cross-scale topological evolution module is input into the adversarial optimization decision module generator as a gradient distribution constraint boundary; The three-dimensional crack network topology of the cross-scale topological evolution module is input into the adversarial optimization decision module discriminator to drive the adjustment of the crack sensitivity coefficient.
10. The electroluminescent display emitter package structure optimization system of claim 9, wherein, Further comprising: The buffer layer fractal geometric configuration generated by the adversarial optimization decision module is input into the self-verification laser execution module; The femtosecond laser direct writing angle is set according to the fractal dimension range of the fractal geometric configuration, which is 1.2 to 1.8.