A method for optimizing light intensity control of organic electroluminescent film

Abstract

The present application provides an organic electroluminescent film light intensity control optimization method, comprising: obtaining an organic light-emitting diode device sample set including a combination of multiple electrode materials and organic materials, wherein each device sample in the sample set corresponds to an electrode-organic layer combination; using a first-principles calculation method to calculate the energy level distribution of each device sample in the sample set, and obtaining a work function value and an energy level steepness parameter of each device sample; training a machine learning model according to the work function value and the energy level steepness parameter, and establishing a mapping relationship between the electrode material, the organic material and the device energy level distribution; constructing a device interface structure model for the screened energy level mismatch interface combination, and introducing a work function gradient layer between the electrode and the organic layer.

Description

Technical Field

[0001] The invention relates to the field of information technology, and in particular to a method for optimizing light intensity control of an organic electroluminescent film. Background Art

[0002] The performance of organic light-emitting diode devices depends largely on the interface characteristics between the electrode and the organic layer. However, the combination of different electrode materials and organic materials will lead to energy level mismatch, forming a high injection barrier, which seriously affects the charge injection efficiency and luminescence performance of the device. How to systematically study and optimize this interface characteristic has become a key technical problem that needs to be solved urgently. Traditional experimental methods are difficult to fully explore a large number of material combinations, and theoretical calculations often ignore the complex microstructure of the interface. Therefore, it is urgent to establish an efficient method to predict and screen out interface combinations with serious energy level mismatches, and design targeted improvement plans. This requires us to accurately describe the energy level distribution between the electrode and the organic layer, especially the key parameters of work function and energy level steepness. At the same time, it is also necessary to deeply understand the microstructural evolution law of the work function gradient layer and find a material component matching scheme to achieve an ideal work function gradient. This series of challenges involves the coupling of multi-scale physical processes, and requires the organic combination of first-principles calculations, machine learning, and molecular dynamics simulation methods to build a multi-level correlation model from material components to device performance. Summary of the invention

[0003] The present invention provides an organic electroluminescent film light intensity control optimization method, which mainly includes:

[0004] Acquire an organic light-emitting diode device sample set including a plurality of electrode material and organic material combinations, wherein each device sample in the sample set corresponds to an electrode-organic layer combination;

[0005] The energy level distribution of each device sample in the sample set is calculated by using a first-principle calculation method to obtain a work function value and an energy level steepness parameter of each device sample;

[0006] According to the work function value and the energy level steepness parameter, a machine learning model is trained to establish a mapping relationship between electrode materials, organic materials and device energy level distribution;

[0007] Using machine learning models, we predict the work function values ​​and energy level steepness of different electrode-organic layer combinations, and screen out interface combinations with severe energy level mismatch and high injection barriers.

[0008] For the screened energy level mismatch interface combination, a device interface structure model is constructed, and a work function gradient layer is introduced between the electrode and the organic layer;

[0009] The molecular dynamics simulation method is used to study the microstructural evolution law of work function gradient layers with different material compositions and thickness distributions, and to obtain the material composition matching scheme required to achieve the ideal work function gradient.

[0010] As a preferred solution of the method for optimizing the light intensity control of the organic electroluminescent film of the present invention, the method of obtaining a sample set of organic light emitting diode devices including a combination of multiple electrode materials and organic materials, wherein each device sample in the sample set corresponds to an electrode-organic layer combination, comprises:

[0011] Acquire a pre-established set of organic light-emitting diode device samples, each sample in the set of samples including combination information of electrode materials and organic materials;

[0012] For each sample, according to its electrode material and organic material combination, the corresponding organic light-emitting diode device structure is determined using a pre-established device structure database;

[0013] For a determined device structure, obtaining performance parameters of the structure through an electrical performance testing system, wherein the performance parameters include luminous efficiency and color coordinates;

[0014] The obtained performance parameters are compared with the preset thresholds. If the performance parameters meet the preset thresholds, the sample is marked as a qualified sample.

[0015] As a preferred solution of the method for optimizing the light intensity control of the organic electroluminescent film of the present invention, the energy level distribution of each device sample in the sample set is calculated by the first principle calculation method to obtain the work function value and energy level steepness parameter of each device sample, including:

[0016] For each device sample, obtain its material properties, and use density functional theory to calculate the electronic structure and obtain energy level distribution information;

[0017] By statistically analyzing the energy level distribution information and curve fitting, the work function value and energy level steepness parameters are determined;

[0018] The work function value and the energy level steepness parameter are used as input, and the device performance parameter is used as output to train a support vector machine regression model to obtain a device performance prediction model;

[0019] If the predicted performance parameters of the new device sample are better than the preset threshold, the genetic algorithm is used to optimize the device structure parameters and process parameters based on the predicted performance parameters and material properties to obtain a new device design with better performance.

[0020] As a preferred solution of the method for optimizing the light intensity control of the organic electroluminescent film of the present invention, the mapping relationship between the electrode material, the organic material and the device energy level distribution is established according to the work function value and the energy level steepness parameter, including:

[0021] Obtaining work function values ​​and energy level steepness parameters of the electrode material and the organic material, performing preprocessing of the data by removing outliers and normalizing, and using the preprocessed data as input features of the machine learning model;

[0022] According to the energy level distribution characteristics of the device, support vector regression or multi-layer perceptron neural network algorithm is used to establish a mapping relationship model between material performance parameters and device energy level distribution;

[0023] According to the mapping relationship model, the performance parameters of the material to be predicted are input to predict its energy level distribution in the device.

[0024] As a preferred solution of the method for optimizing the light intensity control of the organic electroluminescent film of the present invention, the method uses a machine learning model to predict the work function values ​​and energy level steepness of different electrode-organic layer combinations, and screens out interface combinations with severe energy level mismatch and high injection barriers, including:

[0025] Obtaining electrode work function, organic layer electron affinity and organic layer ionization potential material parameters of candidate electrode-organic layer combinations from a pre-established database of electrode materials and organic layer materials;

[0026] For each candidate electrode-organic layer combination, the material parameters are normalized and key characteristic parameters are extracted;

[0027] If the electrode-organic layer combination has both severe energy level mismatch and high injection barrier, it is output to a CSV file, which includes the electrode material, organic layer material, energy level mismatch degree value and injection barrier height value.

[0028] As a preferred solution of the method for optimizing the light intensity control of the organic electroluminescent film of the present invention, the device interface structure model is constructed for the screened energy level mismatch interface combination, and a work function gradient layer is introduced between the electrode and the organic layer, including:

[0029] Obtaining material parameters and energy level data of the electrode layer and the organic layer, constructing a device interface structure model, and simulating and analyzing the energy level matching and carrier injection and transmission characteristics between the electrode layer and the organic layer;

[0030] If the interface energy level mismatch is serious, a support vector machine algorithm is used to optimize the material composition and thickness of the work function gradient layer so that a continuous work function change is formed between the electrode layer and the organic layer;

[0031] According to the simulation analysis results, the structural parameters of the work function gradient layer are fine-tuned until the best device interface energy level matching and carrier transport characteristics are obtained.

[0032] As a preferred solution of the method for optimizing the light intensity control of the organic electroluminescent film of the present invention, the acquisition of material parameters and energy level data of the electrode layer and the organic layer, the construction of the device interface structure model, and the simulation analysis of the energy level matching and carrier injection and transmission characteristics between the electrode layer and the organic layer specifically include:

[0033] Obtain material parameters of the electrode layer and organic layer and construct a device interface structure model;

[0034] The energy level distribution data on both sides of the interface are obtained by using quantum chemical calculation methods;

[0035] If the interface energy level difference exceeds the preset threshold, it is judged that the energy level mismatch is serious;

[0036] If the energy level difference is lower than the preset threshold, the energy level match is judged to be good;

[0037] According to the results of energy level matching analysis, the Monte Carlo method is used to simulate the injection and transmission process of carriers between interfaces, and the carrier transition probability, average mobility and recombination probability are calculated.

[0038] If the device performance index is lower than the preset requirements, the electrode layer or organic layer material parameters are adjusted to optimize the interface energy level matching and carrier transport characteristics until the device performance meets the requirements.

[0039] As a preferred solution of the method for optimizing the light intensity control of the organic electroluminescent film of the present invention, the molecular dynamics simulation method is used to obtain the microstructure evolution law of the work function gradient layer with different material components and thickness distributions to achieve the material component matching scheme required for the ideal work function gradient, including:

[0040] Using LAMMPS molecular dynamics simulation software, a work function gradient layer model with different material components and thickness distributions was established;

[0041] Set initial parameters, including atomic ratio of material components, thickness distribution data and force field parameters of atomic interactions, and perform geometric optimization and energy minimization on the model to generate a stable initial simulation structure;

[0042] Use LAMMPS software to perform canonical ensemble molecular dynamics simulation calculations and set the number of simulation steps and time step;

[0043] Record the atomic coordinates and force data at different times during the simulation to obtain the trajectory information of microstructure evolution;

[0044] Perform statistical analysis on the trajectory data obtained from the simulation and calculate the work function values ​​under different material compositions and thicknesses;

[0045] Draw a three-dimensional surface diagram of the work function changing with the atomic ratio and thickness distribution of the material components, and intuitively display the distribution law of the work function gradient;

[0046] According to the quantitative indicators of the target work function gradient distribution, which include the work function variation range and gradient variation rate, the matching degree between the simulation results and the target is compared and analyzed;

[0047] Screen out several groups of material components and thickness combinations whose work function gradient distribution is closest to the target as candidate optimization solutions;

[0048] For the selected candidate optimization solutions, molecular dynamics simulations are performed on a longer time scale to evaluate their structural stability under high temperature or stress environments;

[0049] At the same time, the deviations between the work function values ​​of different schemes and the experimental measurements are compared to verify the accuracy of the model;

[0050] If the stability and accuracy evaluation results of the candidate solution meet the preset indicators, it will be determined as the optimal material composition and thickness matching solution;

[0051] If the requirements cannot be met, the material component ratio and thickness distribution are adjusted according to the evaluation results through orthogonal experimental design or machine learning optimization method to generate new candidate solutions, and the simulation verification process is repeated until the optimal solution is obtained;

[0052] The final optimal material composition and thickness matching scheme is output as a detailed report, including the atomic ratio of each component, the thickness value of each layer, the three-dimensional surface diagram of the work function gradient distribution, and the simulation verification results under different environmental conditions, for reference in subsequent material preparation and device application.

[0053] As a preferred solution of the method for optimizing the light intensity control of the organic electroluminescent film of the present invention, the three-dimensional surface diagram of the work function variation with the atomic ratio and thickness distribution of the material components is drawn to intuitively display the distribution law of the work function gradient. According to the quantitative indicators of the target work function gradient distribution, such as the work function variation range and the gradient variation rate, the matching degree between the simulation results and the target is compared and analyzed, and several groups of material components and thickness combinations whose work function gradient distribution is closest to the target are selected as candidate optimization schemes. The selected candidate optimization schemes are subjected to molecular dynamics simulation on a longer time scale to evaluate their structural stability under high temperature or stress environment, specifically including:

[0054] Constructing an initial atomic structure model of the material according to the material components and atomic ratios;

[0055] Performing hierarchical division on the initial atomic structure model to obtain a hierarchical atomic structure model;

[0056] Performing first-principles calculations on the layered atomic structure model to obtain work function values ​​of each layer;

[0057] Determining a gradient variation range and a variation rate of the work function according to the work function value;

[0058] Molecular dynamics simulation method is used to simulate the electron emission process of the material under the action of an external electric field, and the time evolution and spatial distribution results of the electron emission are obtained;

[0059] Comparing the electron emission current density with the experimental measurement value, if the relative error is lower than a preset threshold, taking the current material composition and atomic ratio as a candidate solution;

[0060] The candidate solutions are subjected to nanosecond-scale molecular dynamics simulations to evaluate their structural stability under high temperature conditions, and the one with the most stable structure is selected as the optimal solution.

[0061] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0062] The present invention discloses a method for optimizing the energy level of an organic light-emitting diode device. The method constructs a device sample set including a combination of multiple electrodes and organic materials, obtains the work function and energy level steepness parameters of each sample by first-principles calculation, and trains a machine learning model to establish a mapping relationship between the material and the energy level distribution. The model is used to predict and screen out interface combinations with severe energy level mismatch, and a work function gradient layer is introduced in a targeted manner. The microstructural evolution of the gradient layer is studied by molecular dynamics simulation, and the material component scheme for achieving an ideal work function gradient is determined. The present invention realizes the accurate prediction and optimized design of the energy level distribution of the organic light-emitting diode device, effectively reduces the interface injection barrier, and improves the device performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 The present invention is a flow chart of an organic electroluminescent film light intensity control optimization method.

[0064] Figure 2 It is a schematic diagram of a method for optimizing light intensity control of an organic electroluminescent film according to the present invention.

[0065] Figure 3 It is another schematic diagram of a method for optimizing light intensity control of an organic electroluminescent film according to the present invention. DETAILED DESCRIPTION

[0066] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0067] like Figure 1-Figure 3 In this embodiment, a method for optimizing light intensity control of an organic electroluminescent film may specifically include:

[0068] S101 , obtaining an organic light emitting diode device sample set including a plurality of electrode material and organic material combinations, wherein each device sample in the sample set corresponds to an electrode-organic layer combination.

[0069] A pre-established set of organic light-emitting diode device samples is obtained, wherein each sample in the sample set includes combination information of electrode materials and organic materials; for each sample, the corresponding organic light-emitting diode device structure is determined according to its electrode material and organic material combination using a pre-established device structure database; for the determined device structure, performance parameters of the structure are obtained through an electrical performance test system, wherein the performance parameters include luminous efficiency and color coordinates; the obtained performance parameters are compared with a preset threshold value, and if the performance parameters meet the preset threshold value, the sample is marked as a qualified sample.

[0070] Exemplarily, a pre-established set of organic light-emitting diode device samples is obtained, wherein each sample includes combination information of electrode materials and organic materials. Each sample is selected from the sample set in turn to obtain the electrode material and organic material combination corresponding to the sample. According to the obtained electrode material and organic material combination, the organic light-emitting diode device structure corresponding to the sample is determined using a pre-established device structure database. For the determined organic light-emitting diode device structure, the performance parameters of the structure, including luminous efficiency and color coordinates, are obtained through an electrical performance test system. The obtained performance parameters are compared with a preset threshold, wherein the preset threshold is determined based on the device application requirements using the performance parameters of a large number of qualified devices. If the performance parameters meet the preset threshold, the sample is marked as a qualified sample. The number of qualified samples in the currently tested samples is counted. If the number of qualified samples reaches a preset number (the preset number is determined using a statistical method based on the confidence requirements), the currently tested sample subset is determined as a sample set that meets the requirements, and the traversal is terminated; if the number of qualified samples does not reach the preset number, the next sample is selected from the untested samples for processing, and the above steps are repeated until all samples are traversed and tested.

[0071] S102, using a first-principles calculation method to calculate the energy level distribution of each device sample in the sample set, to obtain a work function value and an energy level steepness parameter of each device sample.

[0072] For each device sample, its material properties are obtained, and the electronic structure is calculated using density functional theory to obtain energy level distribution information. The work function value and energy level steepness parameter are determined by statistical analysis and curve fitting of the energy level distribution information. The work function value and energy level steepness parameter are used as input, and the device performance parameters are used as output to train the support vector machine regression model to obtain a device performance prediction model. If the predicted performance parameters of the new device sample are better than the preset threshold, the device structure parameters and process parameters are optimized using a genetic algorithm based on the predicted performance parameters and material properties to obtain a new device design with better performance.

[0073] Exemplarily, according to the material properties of the device sample, a first-principle calculation method based on quantum mechanics, such as density functional theory (DFT), is used to calculate the electronic structure of each device sample to obtain the energy level distribution information of the device sample. By performing statistical analysis and curve fitting on the energy level distribution information of the device sample, the work function value and energy level steepness parameter of each device sample are extracted. The work function value reflects the minimum energy required for electrons to escape from the device surface, and the energy level steepness parameter characterizes the steepness of the device energy level distribution. According to the work function value and the energy level steepness parameter, a physical model or an empirical formula is used to calculate the electron emission performance and current transfer characteristics of the device sample to obtain the performance parameters of the device sample, such as the electron emission threshold electric field strength and the emission current density. The work function value, energy level steepness parameter and performance parameter of the device sample are normalized to construct a feature vector. The principal component analysis (PCA) method is used for feature selection and dimensionality reduction to obtain an optimized feature vector. With the optimized feature vector as input and the device performance parameter as output, a support vector machine (SVM) regression model is trained to obtain a device performance prediction model. Using the trained device performance prediction model, the feature vectors of new device samples are predicted to obtain the predicted performance parameters of the new device samples. The predicted performance parameters are compared with the preset performance thresholds set according to the device application requirements. If the predicted performance parameters are better than the preset performance thresholds, the new device samples are judged to meet the performance requirements. According to the predicted performance parameters and material properties of the new device samples, parameter optimization algorithms such as genetic algorithms and particle swarm optimization algorithms are used to optimize the device's structural parameters (such as gate length, channel doping concentration) and process parameters (such as oxide layer thickness, annealing temperature) to obtain a new device design with better performance. Device simulation is performed on the optimized new device design to evaluate its performance improvement effect. At the same time, the holdout method or cross-validation method is used to evaluate the generalization performance of the device performance prediction model, and the model is tuned and improved based on the evaluation results to improve the accuracy and reliability of device performance prediction.

[0074] S103. According to the work function value and the energy level steepness parameter, a machine learning model is trained to establish a mapping relationship between electrode materials, organic materials and device energy level distribution.

[0075] The work function values ​​and energy level steepness parameters of the electrode material and the organic material are obtained, and the data is preprocessed to remove outliers and normalize, and the preprocessed data is used as the input features of the machine learning model; according to the energy level distribution characteristics of the device, support vector regression or multi-layer perceptron neural network algorithm is used to establish a mapping relationship model between the material performance parameters and the device energy level distribution; according to the mapping relationship model, the performance parameters of the material to be predicted are input to predict its energy level distribution in the device.

[0076] Exemplarily, the work function values ​​and energy level steepness parameters of electrode materials and organic materials are obtained, and the data is preprocessed, including removing outliers and normalizing the data, and the processed data is used as the input features of the machine learning model. According to the characteristics of the device energy level distribution, a machine learning algorithm suitable for processing continuous data is selected, such as support vector regression (SVR) or multi-layer perceptron neural network (MLP). The reason for selecting these two types of algorithms is that they can fit nonlinear relationships and are suitable for modeling complex mappings between material properties and device energy level distribution. The obtained material performance parameters and the corresponding device energy level distribution data are randomly divided into training sets, validation sets, and test sets, and the ratio can be 7:2:1. The model is trained on the training set, and the model performance is evaluated on the validation set. According to the evaluation results, the hyperparameters of the model are adjusted, such as the kernel function type and regularization coefficient of SVR, or the number of hidden layers, number of neurons, and activation function type of MLP. Repeat the training and parameter adjustment process until satisfactory performance is obtained on the validation set. Use the trained model to evaluate performance on the test set, calculate the mean square error (MSE) and mean absolute error (MAE) indicators, and judge the generalization ability of the model. According to the established mapping relationship model, input the work function value and energy level steepness parameter of the material to be predicted to predict its energy level distribution in the device. Compare the prediction results with the device performance requirements, such as carrier mobility and switching ratio, to determine whether the material meets the requirements. If it does, it is considered that the material is suitable for the preparation of the corresponding device; otherwise, it is necessary to adjust the material properties or select other materials. By increasing the number of training samples, optimizing the model structure and hyperparameters, the accuracy of the prediction of the correlation between material properties and device performance is improved. At the same time, combined with domain knowledge and physical mechanisms, the prediction results are analyzed and interpreted to provide guidance for material screening and device design.

[0077] S104. Use machine learning models to predict the work function values ​​and energy level steepness of different electrode-organic layer combinations, and screen out interface combinations with severe energy level mismatch and high injection barriers.

[0078] From a pre-established database of electrode materials and organic layer materials, the electrode work function, organic layer electron affinity and organic layer ionization potential material parameters of the candidate electrode-organic layer combination are obtained; for each candidate electrode-organic layer combination, the material parameters are normalized and key characteristic parameters are extracted; if the electrode-organic layer combination has both severe energy level mismatch and high injection barrier, they are output to a CSV file, which includes electrode materials, organic layer materials, energy level mismatch degree values ​​and injection barrier height values.

[0079] Exemplarily, according to a pre-established database of electrode materials and organic layer materials, the material parameters of the candidate electrode-organic layer combination are obtained, including the electrode work function, the organic layer electron affinity, and the organic layer ionization potential. For each candidate electrode-organic layer combination, the material parameters are normalized and the key characteristic parameters are extracted. The work function value of each combination is predicted using the pre-trained support vector machine model with the extracted characteristic parameters as input. For each candidate electrode-organic layer combination, the normalized material parameters are input into the pre-trained multi-layer perceptron neural network model to predict the steepness of its energy level, that is, the difference between the electrode Fermi level and the HOMO energy level and LUMO energy level of the organic layer. According to the predicted work function value and energy level steepness, the energy level mismatch degree of each candidate electrode-organic layer combination is calculated. Energy level mismatch degree = electrode work function - organic layer ionization potential (for hole injection) or organic layer electron affinity - electrode work function (for electron injection). The calculated energy level mismatch degree is compared with the preset mismatch severity threshold (such as 3eV). If it is greater than the threshold, it is judged that the electrode-organic layer combination has a serious energy level mismatch. For the electrode-organic layer combination with serious energy level mismatch, calculate its injection barrier height. Injection barrier height = electrode work function - organic layer HOMO energy level (for hole injection) or organic layer LUMO energy level - electrode work function (for electron injection). The injection barrier height is compared with the preset high injection barrier threshold (such as 4eV). If it is greater than the threshold, it is judged that the electrode-organic layer combination has a high injection barrier. The electrode-organic layer interface combination with both severe energy level mismatch and high injection barrier obtained by screening is output to a CSV file, including electrode materials, organic layer materials, energy level mismatch degree values ​​and injection barrier height values.

[0080] S105. Construct a device interface structure model for the screened energy level mismatch interface combination, and introduce a work function gradient layer between the electrode and the organic layer.

[0081] The material parameters and energy level data of the electrode layer and the organic layer are obtained, a device interface structure model is constructed, and the energy level matching and carrier injection and transfer characteristics between the electrode layer and the organic layer are simulated and analyzed; if the interface energy level mismatch is serious, a support vector machine algorithm is used to optimize the material composition and thickness of the work function gradient layer so that a continuous work function change is formed between the electrode layer and the organic layer; according to the simulation and analysis results, the structural parameters of the work function gradient layer are fine-tuned until the best device interface energy level matching and carrier transfer characteristics are obtained.

[0082] Based on the material parameters and energy level data of the electrode layer and the organic layer, a device interface structure model is constructed. By simulating and analyzing the energy level matching and carrier injection and transmission characteristics between the electrode layer and the organic layer, it is determined whether the interface energy level mismatch is serious.

[0083] Obtain the material parameters of the electrode layer and the organic layer, and construct the device interface structure model; use quantum chemical calculation methods to obtain the energy level distribution data on both sides of the interface. If the interface energy level difference exceeds the preset threshold, it is judged that the energy level mismatch is serious; if the energy level difference is lower than the preset threshold, it is judged that the energy level matching is good. According to the results of the energy level matching analysis, the Monte Carlo method is used to simulate the injection and transmission process of carriers between interfaces, and the carrier transition probability, average mobility and recombination probability are calculated. If the device performance index is lower than the preset requirements, adjust the electrode layer or organic layer material parameters to optimize the interface energy level matching and carrier transmission characteristics until the device performance meets the standard.

[0084] Exemplarily, according to the material parameters of the electrode layer and the organic layer, a device interface structure model is constructed, and the energy level distribution data on both sides of the interface are obtained by quantum chemical calculation. Using the obtained energy level distribution data, the first principle calculation method, such as density functional theory (DFT), is used to simulate and analyze the energy level matching between the electrode layer and the organic layer, and the energy level difference between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) at the interface is calculated. The calculated interface energy level difference is compared with the empirical value range. If the energy level difference exceeds the upper limit of the empirical value range, it is judged that the interface energy level mismatch is serious; if the energy level difference is lower than the lower limit of the empirical value range, it is judged that the interface energy level matching is good. Based on the interface energy level matching analysis, the Monte Carlo method is used to simulate the injection and transmission process of carriers between interfaces. By random sampling and statistics, the motion trajectories of a large number of carriers are simulated, the transition probability and average mobility of carriers between the electrode layer and the organic layer are counted, and the recombination probability of carriers is calculated. According to the carrier mobility and recombination rate data obtained by Monte Carlo simulation, the injection efficiency and transmission performance of carriers in the device are judged, and then the key performance indicators of the device's current density and luminous efficiency are evaluated. Compare the device performance indicators with the preset performance requirements. If the device performance is lower than expected, such as insufficient current density and too low luminous efficiency, the device structure needs to be further optimized. In the case where the device performance does not meet the standards, the interface energy level matching and carrier transport characteristics are improved by adjusting the material parameters of the electrode layer or organic layer, such as work function, electron affinity, and carrier mobility. At the same time, the interface energy level difference can be reduced and the carrier injection efficiency can be improved by optimizing the interface structure, such as introducing an interface modification layer and regulating the interface dipole mode. After the material and structure are adjusted, the interface energy level matching analysis and carrier transport simulation are re-performed to evaluate the optimization effect until the device performance meets the preset requirements. Finally, based on the results of interface energy level matching analysis and carrier transport simulation, we select electrodes and organic material combinations with good energy level matching, excellent carrier injection and transport performance, and optimize the interface structure design, such as controlling work function matching and introducing interface dipole layers, to obtain efficient and stable devices. Through systematic material screening, interface energy level regulation and carrier transport optimization, we can guide device structure design and improve the photoelectric conversion efficiency and service life of the device.

[0085] S106. Use molecular dynamics simulation methods to study the microstructural evolution of work function gradient layers with different material compositions and thickness distributions, and obtain the material composition matching scheme required to achieve an ideal work function gradient.

[0086] The LAMMPS molecular dynamics simulation software was used to establish a work function gradient layer model with different material components and thickness distributions. The initial parameters were set, including the atomic ratio of material components, thickness distribution data, and the parameters of the force field of atomic interaction, and the model was geometrically optimized and energy minimized to generate a stable initial simulation structure. The canonical ensemble molecular dynamics simulation calculation was performed using the LAMMPS software, and the number of simulation steps and time step were set. The atomic coordinates and force data at different times during the simulation were recorded to obtain the trajectory information of the microstructure evolution. The trajectory data obtained by the simulation were statistically analyzed to calculate the work function values ​​under different material components and thicknesses. A three-dimensional surface diagram of the work function with the atomic ratio of material components and thickness distribution was drawn to intuitively display the distribution law of the work function gradient. According to the quantitative indicators of the target work function gradient distribution, such as the work function variation range and the gradient variation rate, the matching degree between the simulation results and the target was compared and analyzed. Several groups of material components and thickness combinations with the work function gradient distribution closest to the target were selected as candidate optimization schemes. For the selected candidate optimization schemes, molecular dynamics simulations of longer time scales were performed to evaluate their structural stability under high temperature or stress environments. At the same time, the deviations between the work function values ​​of different schemes and the experimental measurements are compared to verify the accuracy of the model. If the stability and accuracy evaluation results of the candidate scheme meet the preset indicators, it is determined as the optimal material component and thickness matching scheme. If the requirements cannot be met, the material component ratio and thickness distribution are adjusted according to the evaluation results through orthogonal experimental design or machine learning optimization method, and a new candidate scheme is generated. The simulation verification process is repeated until the optimal solution is obtained. The final optimal material component and thickness matching scheme is output as a detailed report, including the atomic ratio of each component, the thickness value of each layer, the three-dimensional surface diagram of the work function gradient distribution, and the simulation verification results under different environmental conditions, for reference in subsequent material preparation and device application.

[0087] Exemplarily, LAMMPS molecular dynamics simulation software is used to establish a work function gradient layer model with different material components and thickness distributions. Set initial parameters, including the atomic ratio of material components, thickness distribution data, and the force field parameters of the interaction between atoms, and perform geometric optimization and energy minimization on the model to generate a stable initial simulation structure. Use LAMMPS software to perform canonical ensemble molecular dynamics simulation calculations, set the number of simulation steps and time step. Record the atomic coordinates and force data at different times during the simulation process to obtain the trajectory information of the microstructure evolution. Perform statistical analysis on the trajectory data obtained by the simulation, and calculate the work function values ​​under different material components and thicknesses. Draw a three-dimensional surface diagram of the work function with the atomic ratio and thickness distribution of the material components to intuitively display the distribution law of the work function gradient. According to the quantitative indicators of the target work function gradient distribution, such as the work function variation range and the gradient variation rate, compare and analyze the degree of matching between the simulation results and the target. Screen out several groups of material components and thickness combinations with the work function gradient distribution closest to the target as candidate optimization schemes. For the selected candidate optimization schemes, molecular dynamics simulations of longer time scales are performed to evaluate their structural stability under high temperature or stress environments. At the same time, the deviations between the work function values ​​of different schemes and the experimental measurements are compared to verify the accuracy of the model. If the stability and accuracy evaluation results of the candidate scheme meet the preset indicators, it is determined as the optimal material component and thickness matching scheme. If the requirements cannot be met, the material component ratio and thickness distribution are adjusted according to the evaluation results through orthogonal experimental design or machine learning optimization method, and a new candidate scheme is generated. The simulation verification process is repeated until the optimal solution is obtained. The final optimal material component and thickness matching scheme is output as a detailed report, including the atomic ratio of each component, the thickness value of each layer, the three-dimensional surface diagram of the work function gradient distribution, and the simulation verification results under different environmental conditions, for reference in subsequent material preparation and device application.

[0088] The range and rate of work function gradient change are obtained based on the atomic ratio and thickness distribution of material components. The candidate solutions are judged by the degree of match between the molecular dynamics simulation results and the target. A longer time scale simulation is used to evaluate the structural stability under high temperature stress and determine the optimal solution.

[0089] According to the material components and atomic ratios, an initial atomic structure model of the material is constructed; the initial atomic structure model is divided into layers to obtain a layered atomic structure model; the layered atomic structure model is calculated by first principles to obtain the work function values ​​of each layer; according to the work function values, the gradient change range and change rate of the work function are determined; a molecular dynamics simulation method is used to simulate the electron emission process of the material under the action of an external electric field to obtain the time evolution and spatial distribution results of the electron emission; the electron emission current density is compared with the experimental measurement value, and if the relative error is lower than a preset threshold, the current material component and atomic ratio are used as candidate solutions; the candidate solutions are simulated at the nanosecond level of molecular dynamics to evaluate their structural stability under high temperature conditions, and the most stable structure is selected as the optimal solution.

[0090] Exemplarily, according to the material composition and atomic ratio, the BuildCrystals module in the Materials Studio software is used to build the initial atomic structure model of the material. According to the thickness distribution, the Layer Builder module in the Materials Studio is used to divide the atomic structure model into layers to obtain a layered atomic structure model. Using the CASTEP module in the Materials Studio, based on the density functional theory (DFT), the first principle calculation of the layered atomic structure model is performed to obtain the work function value of each layer. According to the work function values ​​of the adjacent layers, the gradient change range and change rate of the work function in the vertical direction are calculated. According to the gradient change range and change rate of the work function, the LAMMPS software is used to perform molecular dynamics simulation to simulate the electron emission process of the material under the action of an external electric field, and the time evolution and spatial distribution results of the electron emission are obtained. The simulated electron emission current density is compared with the target value measured by the experiment, and the relative error between the two is calculated. If the relative error is lower than the preset threshold of 5%, it is considered that the match is high, and the current material composition and atomic ratio are used as candidate solutions; if the relative error is higher than 5%, it is necessary to adjust the atomic ratio and return to step 1 to rebuild the atomic structure model. LAMMPS software is used to perform molecular dynamics simulations on a longer time scale (such as nanoseconds) to evaluate the structural stability of candidate solutions under high temperature (such as 1000K). The radial distribution function and mean square displacement index are used to quantitatively evaluate the degree of change in the material structure. Based on the results of the structural stability evaluation, the most structurally stable solution is selected from the candidate solutions as the optimal solution.

[0091] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing light intensity control of an organic electroluminescent film, characterized in that: The method comprises: Acquire a sample set of organic light-emitting diode devices including a plurality of electrode material and organic material combinations, wherein each device sample in the sample set corresponds to an electrode-organic layer combination; Using a first-principles calculation method, the energy level distribution of each device sample in the sample set is calculated to obtain a work function value and an energy level steepness parameter of each device sample; According to the work function value and the energy level steepness parameter, a mapping relationship between the electrode material, the organic material and the device energy level distribution is established; Using machine learning models, we predict the work function values ​​and energy level steepness of different electrode-organic layer combinations to screen out interface combinations with severe energy level mismatch and high injection barriers. For the screened energy level mismatch interface combination, a device interface structure model is constructed, and a work function gradient layer is introduced between the electrode and the organic layer; Molecular dynamics simulation method is used to obtain the microstructure evolution law of work function gradient layers with different material components and thickness distributions, so as to realize the material component matching scheme required for the ideal work function gradient; The step of obtaining a sample set of organic light-emitting diode devices including a plurality of electrode materials and organic material combinations, wherein each device sample in the sample set corresponds to an electrode-organic layer combination, comprises: Acquire a pre-established set of organic light-emitting diode device samples, each sample in the set of samples including combination information of electrode materials and organic materials; For each sample, according to its electrode material and organic material combination, the corresponding organic light-emitting diode device structure is determined using a pre-established device structure database; For a determined device structure, obtaining performance parameters of the structure through an electrical performance testing system, wherein the performance parameters include luminous efficiency and color coordinates; The obtained performance parameter is compared with a preset threshold value, and if the performance parameter meets the preset threshold value, the sample is marked as a qualified sample; The first principle calculation method is used to calculate the energy level distribution of each device sample in the sample set to obtain the work function value and energy level steepness parameter of each device sample, including: For each device sample, obtain its material properties, and use density functional theory to calculate the electronic structure and obtain energy level distribution information; By statistically analyzing the energy level distribution information and curve fitting, the work function value and energy level steepness parameters are determined; The work function value and the energy level steepness parameter are used as input, and the device performance parameter is used as output to train a support vector machine regression model to obtain a device performance prediction model; If the predicted performance parameters of the new device sample are better than the preset threshold, the genetic algorithm is used to optimize the device structure parameters and process parameters based on the predicted performance parameters and material properties to obtain a new device design with better performance.

2. The method according to claim 1, characterized in that The establishing of a mapping relationship between electrode materials, organic materials and device energy level distribution according to the work function value and the energy level steepness parameter comprises: Obtaining work function values ​​and energy level steepness parameters of the electrode material and the organic material, performing preprocessing of the data by removing outliers and normalizing, and using the preprocessed data as input features of the machine learning model; According to the energy level distribution characteristics of the device, support vector regression or multi-layer perceptron neural network algorithm is used to establish a mapping relationship model between material performance parameters and device energy level distribution; According to the mapping relationship model, the performance parameters of the material to be predicted are input to predict its energy level distribution in the device.

3. The method according to claim 1, characterized in that The machine learning model is used to predict the work function values ​​and energy level steepness of different electrode-organic layer combinations, and screen out interface combinations with severe energy level mismatch and high injection barriers, including: Obtaining the electrode work function, organic layer electron affinity and organic layer ionization potential material parameters of the candidate electrode-organic layer combination from a pre-established database of electrode materials and organic layer materials; For each candidate electrode-organic layer combination, the material parameters are normalized and key characteristic parameters are extracted; If the electrode-organic layer combination has both severe energy level mismatch and high injection barrier, it is output to a CSV file, which includes the electrode material, organic layer material, energy level mismatch degree value and injection barrier height value.

4. The method according to claim 1, characterized in that: The device interface structure model is constructed for the screened energy level mismatch interface combination, and a work function gradient layer is introduced between the electrode and the organic layer, including: Obtaining material parameters and energy level data of the electrode layer and the organic layer, constructing a device interface structure model, and simulating and analyzing the energy level matching and carrier injection and transmission characteristics between the electrode layer and the organic layer; If the interface energy level mismatch is serious, a support vector machine algorithm is used to optimize the material composition and thickness of the work function gradient layer so that a continuous work function change is formed between the electrode layer and the organic layer; According to the simulation analysis results, the structural parameters of the work function gradient layer are fine-tuned until the best device interface energy level matching and carrier transport characteristics are obtained.

5. The method according to claim 4, characterized in that The obtaining of material parameters and energy level data of the electrode layer and the organic layer, constructing a device interface structure model, and simulating and analyzing the energy level matching and carrier injection and transmission characteristics between the electrode layer and the organic layer specifically includes: Obtain material parameters of the electrode layer and organic layer and construct a device interface structure model; The energy level distribution data on both sides of the interface are obtained by using quantum chemical calculation methods; If the interface energy level difference exceeds the preset threshold, it is judged that the energy level mismatch is serious; If the energy level difference is lower than the preset threshold, the energy level match is judged to be good; According to the results of energy level matching analysis, the Monte Carlo method is used to simulate the injection and transmission process of carriers between interfaces, and the carrier transition probability, average mobility and recombination probability are calculated. If the device performance index is lower than the preset requirements, the electrode layer or organic layer material parameters are adjusted to optimize the interface energy level matching and carrier transport characteristics until the device performance meets the requirements.

6. The method according to claim 1, characterized in that The molecular dynamics simulation method is used to obtain the microstructure evolution law of the work function gradient layer with different material components and thickness distributions, so as to realize the material component matching scheme required for the ideal work function gradient, including: Using LAMMPS molecular dynamics simulation software, a work function gradient layer model with different material components and thickness distributions was established; Set initial parameters, including atomic ratio of material components, thickness distribution data and force field parameters of atomic interactions, and perform geometric optimization and energy minimization on the model to generate a stable initial simulation structure; Use LAMMPS software to perform canonical ensemble molecular dynamics simulation calculations and set the number of simulation steps and time step; Record the atomic coordinates and force data at different times during the simulation to obtain the trajectory information of microstructure evolution; Perform statistical analysis on the trajectory data obtained from the simulation and calculate the work function values ​​under different material compositions and thicknesses; Draw a three-dimensional surface diagram of the work function changing with the atomic ratio and thickness distribution of the material components, and intuitively display the distribution law of the work function gradient; According to the quantitative indicators of the target work function gradient distribution, which include the work function variation range and gradient variation rate, the matching degree between the simulation results and the target is compared and analyzed; Screen out several groups of material components and thickness combinations whose work function gradient distribution is closest to the target as candidate optimization solutions; For the selected candidate optimization solutions, molecular dynamics simulations are performed on a longer time scale to evaluate their structural stability under high temperature or stress environments; At the same time, the deviations between the work function values ​​of different schemes and the experimental measurements are compared to verify the accuracy of the model; If the stability and accuracy evaluation results of the candidate solution meet the preset indicators, it will be determined as the optimal material composition and thickness matching solution; If the requirements cannot be met, the material component ratio and thickness distribution are adjusted according to the evaluation results through orthogonal experimental design or machine learning optimization method to generate new candidate solutions, and the simulation verification process is repeated until the optimal solution is obtained; The final optimal material composition and thickness matching scheme is output as a detailed report, including the atomic ratio of each component, the thickness value of each layer, the three-dimensional surface diagram of the work function gradient distribution, and the simulation verification results under different environmental conditions, for reference in subsequent material preparation and device application.

7. The method according to claim 6, characterized in that The three-dimensional surface diagram of work function variation with atomic ratio and thickness distribution of material components is drawn to intuitively display the distribution law of work function gradient. According to the quantitative indicators of target work function gradient distribution, including work function variation range and gradient variation rate, the matching degree between simulation results and target is compared and analyzed, and several groups of material components and thickness combinations whose work function gradient distribution is closest to the target are selected as candidate optimization schemes. The selected candidate optimization schemes are subjected to molecular dynamics simulation on a longer time scale to evaluate their structural stability under high temperature or stress environment, specifically including: Constructing an initial atomic structure model of the material according to the material components and atomic ratios; Performing hierarchical division on the initial atomic structure model to obtain a hierarchical atomic structure model; Performing first-principles calculations on the layered atomic structure model to obtain work function values ​​of each layer; Determining a gradient variation range and a variation rate of the work function according to the work function value; Molecular dynamics simulation method is used to simulate the electron emission process of the material under the action of an external electric field, and the time evolution and spatial distribution results of the electron emission are obtained; Comparing the electron emission current density with the experimental measurement value, if the relative error is lower than a preset threshold, taking the current material composition and atomic ratio as a candidate solution; The candidate solutions are subjected to nanosecond-scale molecular dynamics simulations to evaluate their structural stability under high temperature conditions, and the one with the most stable structure is selected as the optimal solution.

Images