TW201314474A - Optimized analysis method of photoelectric element process parameters - Google Patents

Abstract

This invention provides an optimized analysis method of photoelectric element process parameters, which employs Taguchi method to plan experimental design and uses statistical test method (analysis of variance and regression analysis) in the analysis of experimental data to generate regression equations corresponding to multiple quality characteristics. Finally, residual analysis and normality test are used to confirm that each regression equation can give an optimized combination of process parameters corresponding to the quality characteristics of the photoelectric element.

Description

Optimized analysis method for process parameters of photoelectric components

The invention is related to the design planning of industrial process parameters, in particular to an optimized analysis method for the process parameters of the photoelectric components by means of statistical verification.

In the production process of industrial products, there are various possible reasons for influencing the characteristics of products, generally referred to as quality characteristics; the influencing factors are called factors, the relevant parameters of each factor are called level, and the process results of the products are determined by The combination of various levels of factors is intended to find some of the most influential factors, and the level of the specific factors can be changed to achieve the desired results. In a variety of experimental planning, the Taguchi Law Institute researches and develops various orthogonal tables, dotted lines, application techniques and analytical methods, and pursues the goal of manufacturing the highest quality products at the lowest cost. In terms of R&D costs, the Taguchi method shortens development time and reduces Resource use is considered. In terms of manufacturing cost, the Taguchi method uses low-grade raw materials and inexpensive equipment to maintain a certain level of quality. To this end, the experimental planning of the Taguchi method is the most widely used in process engineering research in various industrial fields.

With the rapid development of the optoelectronic industry, the green environmental protection panel has begun to change. In recent years, many academic research teams and the optoelectronic industry have mostly bet on emerging process products such as organic light emitting diodes (OLEDs). Many efforts have been made to develop and improve; however, in the development of structures or materials of emerging photovoltaic components, it is necessary to consider the physical properties of the materials themselves, such as energy level difference, carrier mobility, thermal properties, morphology, and the like. Taking an OLED device as an example, whether the thickness of the hole injection layer, the thickness of the hole transport layer, the thickness of the light-emitting layer, the concentration of the guest dopant, or the thickness of the electron transport layer, the parameters of any process structure usually directly affect the photoelectric characteristics of the device. . Therefore, the "Studies on modifications of ITO surfaces in OLED devices by Taguchi methods" published by the research team of Jia-Ming Liu et al., Materials Science & Engineering B85, 2001, 209-211, uses the Taguchi method to study OLED components. Process ITO coating is affected by various surface treatments, including chemicals and mechanical processes under different process conditions; the team uses the Taguchi method to analyze the best surface treatment conditions, and then use the best condition process OLED The luminous efficiency of the component can be increased by more than 50%.

Even in the optoelectronics industry, products with quite mature technologies, such as liquid crystal displays, due to the fierce competition of global panel factories, production capacity, quality and price are the iron laws of products. Therefore, in addition to actively increasing production capacity, it is necessary to target products. The correlation of various quality characteristics is discussed in depth, so that the lowest experimental cost can fully consider all the quality characteristics to enable the development of the most competitive display products. Therefore, as in the case of Jiang Jizhe's 2008 Master Kaohsiung University Master's thesis, "The study of the fuzzy Taguchi method in the process of multi-quality characteristics - taking the liquid crystal display process as an example" is in the metallization process of thin film transistors in the liquid crystal display process. Using the method of processing single quality characteristics in Taguchi method to find the best parameter combination, and then multivariate analysis to find the significant factors that affect the two quality characteristics at the same time, and carry out simulation verification of the best parameter combination including two or more quality characteristics; The numerical analysis of the optimized procedures and modes can verify the validity and practicability of this method, help save the experiment cost, and shorten the time course of the new product from the experimental stage to the production stage.

In view of this, the inventors are more active in emerging optoelectronic component products, and can find the optimal process conditions in an efficient and rapid manner with the lowest experimental cost, taking into account multiple quality characteristics.

The main object of the present invention is to provide an optimized analysis method for the process parameters of photovoltaic components, which can effectively and quickly find the optimum process parameters with the lowest experimental cost, so that the process products can take into account multiple quality characteristics.

In order to achieve the above object, the present invention provides an optimized analysis method for a process parameter of a photovoltaic element comprising the following steps:

a. selecting a plurality of quality characteristics of the above-mentioned photovoltaic elements, listing a majority of control factors affecting the quality characteristics, and setting a variation parameter of each of the control factors;

b. Using the Taguchi method to carry out experimental planning design and parameter configuration, and establish a straight-through table to form a multi-group configuration structure;

c. measuring the quality characteristics of each group of configuration structures, generating a majority of corresponding objective functions;

d. performing a variance analysis on the objective functions of the quality characteristics respectively, and selecting, among the quality characteristics, a plurality of the control factors having a large effect effect and at least two high-order factors combined by the control factors;

e. generating a regression equation for each of the quality characteristics, and using residual analysis to test the difference between the predicted value of each of the quality characteristics of each set of configuration equations and the objective function obtained in step c;

f. The residual distribution of the normal state verification step e is a normal state distribution, that is, each regression equation can be determined to obtain a process parameter optimization combination corresponding to each quality characteristic of the photovoltaic element.

In the method for optimizing the process parameters of the photovoltaic device provided by the preferred embodiment of the present invention, the photovoltaic element is made of a plurality of organic materials, wherein one layer of the organic material is a light-emitting layer of an electron and a hole recombination zone. In a, the control factors include the thickness of each layer of organic material and the dopant concentration of the luminescent layer.

In the method for optimizing the process parameters of the photovoltaic device according to the preferred embodiment of the present invention, in step a, the selected quality characteristics include the driving voltage, the luminance, and the luminous efficiency of the photovoltaic device.

Preferably, in step c, the method for measuring each of the quality characteristics is: measuring the driving voltage: providing a plurality of constant currents of the photovoltaic element, and obtaining an operation curve corresponding to the driving voltage, or providing a majority of the photoelectric element Voltage, and obtain the corresponding driving current operating curve, to obtain a driving voltage under a fixed current condition; measurement of luminous brightness: using spectroscopic colorimetric measurement of specific spectral-chromaticity coordinates under the above fixed current conditions Luminous brightness; and measurement of luminous efficiency: the luminous intensity measured by the above-mentioned measured light and the current density corresponding to the fixed current condition are obtained by the following conversion calculation formula: luminous efficiency = luminous brightness / (current density x10).

In the method for optimizing the process parameters of the photovoltaic device provided by the preferred embodiment of the present invention, in step d, each of the control factors and at least one high-order factor pair combined by at least two of the control factors are further evaluated by a statistical hypothesis test. The degree of influence of this quality characteristic.

Preferably, the regression equation generated in step e is Y=β0+β1X1+β2X2+...+βkXk+ε, where X1～Xk are the control factors or high-order factors of the statistical hypothesis test; β1～βk are each control factor Or the coefficient of the high-order factor, the statistical hypothesis is determined to determine whether each coefficient β1 - βk is 0, and if 0, it indicates that the control factor or the high-order factor is a non-significant factor. Better yet, the assumptions for this statistical hypothesis test are:

False hypothesis H0: β1 = β2 = ... = βk = 0

The opposite hypothesis H1: β1, β2, ..., βk≠0

Discard criteria: If P>α, reject H1 (α is the error rate, P is the probability value); the statistical decision hypothesis decision criterion is: If P<α, reject the null hypothesis, if P>>α, then The null hypothesis is correct.

The best one, in step d, is to use the statistical hypothesis of the F-test to determine that α = 0.05 corresponds to reject the null hypothesis or the null hypothesis is correct.

In the optimized analysis method for the process parameters of the photovoltaic element provided by the preferred embodiment of the present invention, after step f, the error of each quality characteristic and the estimated value of the plurality of optimized process parameters is verified by the verification experiment.

Preferably, one of the quality characteristics is a large characteristic, and the optimized process parameters of each group are adapted according to the estimated value.

The invention provides an optimized analysis method for the process parameters of the photoelectric component, and establishes an optimization mode capable of expressing the designer's requirements, including the objective function and the factor condition limit, and optimizes the method to satisfy the constraint condition. The function is the optimal predicted design value. The organic light-emitting diode process parameters are modulated, and the photoelectric characteristics of the components are improved by optimizing parameters to improve the luminous efficiency of the components.

In order to explain the structure, features and advantages of the present invention in detail, the following description of the preferred embodiments and the accompanying drawings, wherein: FIG. The second figure is a structure diagram of an energy band of a photovoltaic element shown in the first figure, which is an organic light-emitting diode; the third figure is a flow chart of a preferred embodiment of the present invention; and the fourth figure is a preferred embodiment of the present invention. For example, the residual voltage distribution of the regression equation of the driving voltage of the photovoltaic element; the fifth figure is the residual distribution of the regression equation of the luminous luminance of the photovoltaic element according to the preferred embodiment of the present invention; and the sixth figure is a preferred embodiment of the present invention. The luminous efficiency of the photoelectric element is returned to the residual distribution of the equation; the seventh figure is a graph of the estimated driving voltage of the three sets of optimized process parameters generated by the preferred embodiment of the present invention; The graph of the predicted illuminance of the three sets of optimized process parameters generated by the embodiment; the ninth graph is the predicted illuminance curve of the three sets of optimized process parameters generated by the preferred embodiment of the present invention. .

Referring to the first embodiment, as shown in the first embodiment, the photoelectric element 1 is analyzed. According to the optimization analysis method of the process parameters provided by the present invention, the photovoltaic element 1 can have an optimum driving voltage and brightness. Quality characteristics such as luminous efficiency. Taking the organic light-emitting diode as an example, the photovoltaic element 1 has a substrate 10, an anode 20, a hole injection layer (HIL) 30, and a hole transport layer stacked sequentially from bottom to top. A hole transport layer (HTL) 40, an emissive layer (EML) 50, an electron transport layer (ETL) 60, an electron injection layer (EIL) 70, and a cathode 80. The light-emitting mechanism of the photovoltaic element 1 is to apply a positive external bias voltage to the anode 20 and the cathode 80, so that the electrons and the holes respectively overcome the interface barrier of the electron injection layer 70 and the hole injection layer 30, respectively. Entering the LOMO (lowest unoccupied molecular orbital) energy level of the electron transport layer 60 and the HOMO (highest occupied molecular orbital) energy level of the hole transport layer 40; when the electrons and holes are recombined in the light-emitting layer 50, an exciton is formed ( Exciton), the excitons migrate by the electric field of the applied bias voltage, and transfer energy to the molecules of the light-emitting layer 50, so that the excited electrons transition from the ground state to the excited state, and the electrons are then radiated from the excited state to the excited state. The ground state releases energy in the form of light or heat.

Among them, the main factors affecting the luminescence characteristics of the organic light-emitting diode are as follows:

(1) Injection efficiency of carriers: To improve the efficiency of hole and electron injection, it is necessary to select electrodes and organic materials with appropriate work functions. As shown in the second figure, the appropriate work function has been selected for the photovoltaic element 1. The material's energy band structure diagram; the material of each layer must be selected to reduce the interface energy barrier between the layers or shorten the barrier width to make the electrons and holes efficient at a fixed voltage. The cathode 80 and the anode 20 inject charge carriers into the organic light-emitting layer 50. The present embodiment further provides an electron injecting layer 70 made of an alkali metal compound, such as an acetate or an alkali metal fluoride, so that the organic light emitting diode lowers the driving voltage and improves the luminous efficiency; the alkali metal compound is steamed in the film. The active metal is cracked and released during the plating process, and LiF is taken as an example, and has an ohmic contact characteristic with the cathode 80 material Al, and as long as a thickness of 1 nm or less is provided between the cathode 80 and the electron transport layer 60, The cathode 80 electrons directly form a penetrating effect from the lowest unoccupied orbital domain of the electron transport layer 60 through the Fermi level of Al, and effectively eliminate the interface energy barrier.

(2) Conductivity of the carrier in the organic film: the organic material such as the hole injection layer 30 and the electron transport layer 60 have a unipolar characteristic, and it is necessary to consider the polarization type of each layer to select a favorable one. Hole transport or a specific material that facilitates electron transport.

(3) Recombination rate of the carrier in the organic film: since the conductivity of the organic material is not the same, and the diffusion distance of the excitons is 20 to 30 nm, the recombined region is usually close to one of the electrodes. Quenching is caused by the proximity of the anode 20 or the cathode 80. Therefore, the organic film of the luminescent layer 50 must have a certain thickness to effectively increase the recombination rate of electrons and holes. The transport layer 60 has a higher highest occupied rail energy level to have a hole blocking capability, and the hole transport layer 40 has a lower minimum unoccupied rail energy level to have an electron blocking capability, and the excitons are limited to The luminescent layer emits light.

(4) Adding a dopant: In a doping system, a high-energy host emitter transfers energy to a low-energy guest emitter dopant, and a guest emitter Luminescence; therefore, the addition of dopants can increase the efficiency of the organic light-emitting diode, and the color of the electroluminescent light can be changed by adding a small amount of the guest emitter, and the efficiency of the organic light-emitting diode can be increased.

Therefore, selecting appropriate materials, increasing the bonding and transmission characteristics between the layers of the photovoltaic element 1, and selecting the hole injection layer 30, the hole transport layer 40, the light-emitting layer 50, and the electrons by the optimized analysis method provided by the present invention. The optimal process parameters of the organic material such as the transport layer 60 not only improve the transmission speed between the electrons and the holes, but also enhance the carrier injection characteristics and the bonding efficiency, so that the photovoltaic element 1 has the best driving voltage and brightness. Quality characteristics such as luminous efficiency. Please refer to the third figure, which is an optimized analysis method for the process parameters of the photovoltaic elements provided by the present invention, and the detailed description steps are as follows:

1. Experimental design: The material structure of each organic layer which is most likely to affect the photoelectric characteristics of the element is set as a control factor. Taking the photovoltaic element 1 provided in this embodiment as an example, the control factor is the hole injection layer 30 (HIL, code A). , hole transport layer 40 (HTL, code B), the main illuminant of the luminescent layer 50 (Host, code C), the guest dopant of the luminescent layer 50 (Dopant, code D) and the electron transport layer 60 (ETL, code name E); and use the process experience to first set the range of variation of each control factor, as shown in the table below, the control factor and variation range of the component structure.

Then, according to the designed control factor and the variation level range, an orthogonal table represented by L _a (b ^c ) is established, where a represents the number of experimental groups, b represents the variation level of each factor, and c represents the number of factors; Whether there is interaction between the numbers and factors, total degrees of freedom, etc., select an appropriate orthogonal table for experimental planning. The process parameters of the L ₁₆ (4 ⁵ ) orthogonal table are designed as shown in the table below.

2. Perform process parameter experiment: According to the established L ₁₆ (4 ⁵ ) orthogonal table process parameters, the process experiments of 16 sets of configuration structures are carried out in sequence, and the measurement results of various quality characteristics of each set of configuration structures are recorded to generate corresponding The objective function, the objective function of the experimental results of each group of different parameter combinations shown in the following table, is the driving voltage V, the luminance L (luminance, cd/m ² ) and the luminous efficiency Y (luminance efficiency) under specific driving current conditions. .

Wherein, the driving voltage V is measured to provide a driving current of a constant current or a constant voltage of the photovoltaic element 1, and an operating curve corresponding to the driving voltage or current is obtained, and the measurement software is measured. The voltage-current data is recorded directly, and a driving voltage is obtained under a fixed current condition (for example, a current density of 50 mA/cm ² ); the measurement of the luminous brightness L is performed by using a spectroscopic color under a driving current of each constant current. A spectrocolorometer measures the illuminance of a specific spectrum-chromatographic coordinate (CIE) and captures the illuminance of the fixed current condition; the luminous efficiency Y is also referred to as current efficiency, and is defined as The number of photons is the ratio of the input charge (electron) in cd/A, which is obtained from the above-mentioned measured luminance L by the following conversion formula:

Luminous efficiency (cd/A) = luminance (cd/m ² ) / (current density (mA/cm ² ) x 10)

3. Analysis of Variance (ANOVA): According to the experimental results in the above table, the variance analysis is performed for each objective function (drive voltage V, luminous brightness L, luminous efficiency Y), and the T test, Z test, card One of the statistical hypothesis tests, such as the square test and the F test, evaluates the degree of influence of each control factor. For example, the F test is used to perform the hypothesis test to test the regression equation Y=β0+β1X1+β2X2+...+βkXk+ In ε, whether the independent variable (X) is a factor with significant influence; the larger the quantitative F value, the smaller the probability from the same sample space, and the higher the influence of this factor. The test method is to determine whether the coefficient β of each variable (X) is 0, and if it is 0, it means that the factor is a non-significant factor. The verification of the statistical hypothesis is to divide the parameters into Null Hypothesis H0 and Altrative Hypothesis H1, which are selected from the null hypothesis and the opposite hypothesis after the verification. The hypothesis is as follows:

False hypothesis H0: β1 = β2 = ... = βk = 0

The opposite hypothesis H1: β1, β2, ..., βk≠0

Discard criteria: If P>α, reject H1

α is the error rate. Because the hypothesis test sometimes fails to fully present the described problem, a standard that can decide to reject or accept the null hypothesis is needed. This standard can be called a significant level or a confidence level α, which is called Error rate; if the significant level α is less than 0.05, it means that there is 95% confidence that the assumed answer is correct, and the error rate is only 5%, so the significant level α is as small as possible.

P is the probability value. If the null hypothesis (H0) is true, then a probability of having the same or more extreme value as the sample statistic can be obtained. There is a great chance that the sampled data can be used as the data. Support evidence that the null hypothesis (H0) is true, that is, evidence that is more intense than the F-test's quantitative F value, within 95% confidence level (α value is 0.05), hypothesis verification of the regression equation , to determine a significant factor. The decision criterion of the P-value method is: If P < α, the null hypothesis (H0) is rejected, and if the P value is >> α, the null hypothesis (H0) is assumed to be correct. Since the present embodiment verifies the null hypothesis with a significant level (α value) of 0.05, the F test table can be obtained as follows:

Driving voltage variation number analysis result: It can be seen from the following table that the detection value of the driving voltage model of the photoelectric element 1 provided in this embodiment is 95% confidence level F _0.05 value can be obtained from the F verification table d1=7, d2=8 The corresponding verification value is 3.50, which is much smaller than the F ₀ value (210.076) in the analysis result, so the null hypothesis (H0) is rejected, indicating that the selected model factor effect is greater than the experimental residual, and the table shows a very small P value (< 0.0001), indicating that the opposite hypothesis (H1) is established, the probability that the model is the same as the residual is less than 0.01%. In addition, the P-value method is used to determine whether each factor is significant. It can be seen from the table that the seven main effect factors include the high-order factors of each control factor A, B, C, D, E, and the combination of the control factors A and C. AC and the higher-order factor DE of the combination of the control factors D and E, wherein the P values of the control factors B, C, E and the higher-order factors AC and DE are smaller than the α value (0.05), so the opposite hypothesis (H1) is established (abandonment criterion) And the P values of the control factors A and D are greater than the alpha value, it should be judged that the null hypothesis (H0) is established as the non-significant factor, but because of the hierarchical principle, a high-order term (high-order factor AC, DE) ) should contain all the low-order terms (control factors A, C, D, E) that make up the high-order term, so the factors A, D can determine that the opposite hypothesis (H1) holds. Through the above analysis, it can be seen that the seven main effect factors A, B, C, D, E, AC, and DE are all significant, and also the effect of these main effect factors on the driving voltage is significant.

Luminescence brightness variation number analysis result: It can be seen from the following table that the detection value F _0.05 value of the luminous brightness model of the photoelectric element 1 provided by the present embodiment at 95% confidence level can be obtained from the F verification table d1=5, d2=10 The corresponding verification value is 3.33, which is less than the F ₀ value of 23.346, so the null hypothesis (H0) is rejected, the selected model factor effect is greater than the experimental residual, and the table shows a very small P value (<0.0001), indicating the opposite hypothesis (H1). When established, the probability of the model being the same as the residual is less than 0.01%. Then, the P value method is used to determine whether each factor is significant. According to the discarding criterion and the hierarchical principle, the five main effect factors include the control factors C, D, E, the higher order factor CE and C ² , so The effect of these five main effect factors on the brightness of the luminescence in the model is significant.

Luminous efficiency variation number analysis result: It can be seen from the following table that the detection value F _0.05 value of the luminous intensity model of the photoelectric element 1 provided by the present embodiment at 95% confidence level can be obtained from the F verification table d1=5, d2=10 The corresponding verification value is 3.33, which is less than the F ₀ value of 24.691, so the null hypothesis (H0) is rejected, the selected model factor effect is greater than the experimental residual, and the table shows a very small P value (<0.0001), indicating that the opposite hypothesis is true ( H1), the probability of the model being the same as the residual is less than 0.01%. Then, the P value method is used to determine whether each factor is significant. According to the discarding criterion and the hierarchical principle, the five main effect factors include the control factors C, D, E, the higher order factor CE and C ² , indicating the illuminance brightness model. The effects of the five main effect factors on luminous efficiency are significant.

4. Regression analysis: By analyzing the experimental results through the analysis of the variance, the significant factors affecting the quality characteristics of the organic light-emitting diode elements are the hole injection layer, the hole transport layer, the electron transport layer, the light-emitting layer and the blending. The impurity concentration and the reaction surface equation of the driving voltage, the illuminating brightness and the luminous efficiency of the organic light emitting diode element can be established as follows:

Driving voltage=-3.803+0.048A+0.042B+0.157C+1.423D+0.146E-1.153×10 ^-3 A*C-0.041D*E

Luminous brightness=-3122.387+220.436C+361.250D+46.557E-1.803C*E-1.54C ²

Luminous efficiency=-6.291+0.445C+0.720D+0.093E-3.654×10 ^-3 C*E-3.108×10 ^-3 C ²

5. Residual analysis: Model Adequacy Checking is performed on the regression equation established after regression analysis, where the definition of Residual is between experimental value and predicted value (Predicted). Difference, the analysis results are shown in the following table, wherein the actual value is the experimental result of the quality characteristics of the photovoltaic element in step 2, and the predicted value is the process parameter of each group of experiments in step 2 is brought into each reaction surface equation respectively The value obtained. Through the analysis of the residuals, we can find the violation of its basic assumptions and the inappropriate model of the model.

6. Normal calibration, draw the normal Probability Plot of the driving voltage, illuminance, and luminous efficiency of the above table, and refer to the fourth to sixth figures, and calculate the average and standard deviation of the analysis data. The number of samples and the value of P are used for verification. If the residual is close to the normal distribution in the normal state test, the pattern formed by the residual value approaches a straight line. It can be seen from the respective graphs that the distribution curves of the residual voltages of the driving voltage, the illuminating luminance, and the luminous efficiency are all in a straight line, so that it can be judged to be close to the normal distribution. In addition, under the condition of 95% confidence level (α=0.05), it can be seen from the analysis data in the figure that the P value (0.847, 0.159, 0.202) of the driving voltage, the illuminating brightness and the luminous efficiency are all larger than the α value (0.05). According to the abandonment criterion, the null hypothesis is established (H0), and the residual is the normal distribution. The regression model constructed in the research is a reasonable and credible model.

7. Optimize the process parameter verification: set the driving voltage in the quality characteristic to the small characteristic, and set the luminous brightness and luminous efficiency to the large characteristic, and adapt the three sets of optimization processes according to the luminous efficiency values from large to small. Parameters, and then calculate the estimated value of the quality characteristics corresponding to each group of process parameters, as shown in the following table, with reference to the seventh to ninth drawings, for the three sets of different process parameters, the driving voltage V, the luminous brightness L and the luminous efficiency Y A graph of each quality characteristic. Then, the component process test is performed to verify the error of each quality characteristic measurement data and the optimized estimation value of each group of process parameters; the verification measurement of each group of process parameters shows that when the organic light emitting diode component structure is Device B-II, when the current density is 50 mA/cm ² , the driving voltage is 8.28 V and the predicted result (9.28 V) is 10.7%, and the luminance is 4701.9 cd/m ² (according to the estimated value error 3 %), luminous efficiency is 9.48 cd/A (error of 3.15% from the estimated value). Therefore, for the brightness and the luminous efficiency of the set-up characteristic, the error of the verification measurement data and the optimization estimation value of the optimized process parameters are all below 10%, even with Device BI and B-II. The verification measurement data obtained from the estimated parameters with the maximum luminous efficiency can be less than 5% and lower than the optimized estimation value.

In summary, the optimal analysis method for the process parameters of the photovoltaic element provided by the present invention is to use the Taguchi method for the experimental design planning, and the statistical verification method (variation number analysis, regression analysis) is used to analyze the experimental data. Corresponding regression equations for most quality characteristics are generated. Finally, residual analysis and normality determination are used to determine the regression equations to obtain the optimal combination of process parameters corresponding to the quality characteristics of the photovoltaic elements. Therefore, when one of the quality characteristics is regarded as a small characteristic, such as a driving voltage, and the rest, such as the luminance of the light or the luminous efficiency, is a large characteristic, the photoelectric element produced by combining the obtained optimal parameters is verified by actually measuring the quality characteristics. It is indeed the structural parameter of the best luminous efficiency of the component.

In summary, the constituent elements and corresponding influence factors disclosed in the foregoing embodiments of the present invention are merely illustrative, and are not intended to limit the scope of the present invention, and the equivalent parameter analysis method of other influencing factors should also be the case. The scope of the patent application is covered and not limited to this.

1. . . Optoelectronic component

10. . . Substrate

20. . . anode

30. . . Hole injection layer

40. . . Hole transport layer

50. . . Luminous layer

60. . . Electronic transport layer

70. . . Electron injection layer

80. . . cathode

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic view of an apparatus for analyzing photovoltaic elements applied to a preferred embodiment of the present invention;

The second figure is an energy band structure diagram of the photoelectric element shown in the first figure, taking an organic light emitting diode as an example;

The third drawing is a flow chart of a preferred embodiment of the present invention;

4 is a residual distribution of a regression equation of a driving voltage of the photovoltaic element according to a preferred embodiment of the present invention;

Figure 5 is a diagram showing the residual distribution of the regression equation of the luminance of the photovoltaic element according to the preferred embodiment of the present invention;

Figure 6 is a diagram showing the residual distribution of the regression equation of the luminous efficiency of the photovoltaic element according to the preferred embodiment of the present invention;

Figure 7 is a graph of estimated drive voltages for three sets of optimized process parameters produced by the preferred embodiment of the present invention;

Figure 8 is a graph showing predicted luminous brightness of three sets of optimized process parameters produced by the preferred embodiment of the present invention;

The ninth graph is a graph of the predicted luminous efficiency of the three sets of optimized process parameters produced by the preferred embodiment of the present invention.

Claims (10)

A method for optimizing the process parameters of a photovoltaic component comprises the steps of: a. selecting a plurality of quality characteristics of the photovoltaic element, listing a majority of control factors affecting the quality characteristics, and setting a variation parameter of each of the control factors; b. Using the Taguchi method to carry out experimental planning design and parameter configuration, establish a straight-through table to form a plurality of configuration structures; c. measure the quality characteristics of each group configuration structure, and generate a majority corresponding objective function; d. respectively The objective function of the quality characteristic is analyzed by the variance number, and each of the quality characteristics selects a plurality of the control factors having a large influence effect and at least two high-order factors combined by the control factor; e. generating a regression equation for each of the quality characteristics, and utilizing The residual analysis tests the difference between the predicted value of each quality characteristic of each set of configuration structures and the objective function obtained in step c; and f. the normal distribution of the residual distribution of step e is normal, that is, determining each The regression equation gives an optimized combination of process parameters for each quality characteristic of the photovoltaic element. The method for optimizing the processing parameters of the photovoltaic device according to claim 1, wherein the photovoltaic element is made of a plurality of organic materials, wherein one layer of the organic material is a light-emitting layer of an electron and a hole recombination zone, and in step a, The control factors include the thickness of each layer of organic material and the dopant concentration of the luminescent layer. The method for optimizing the photoelectric component process parameters according to claim 1, wherein the quality characteristics selected in the step a include a driving voltage, a light emitting luminance, and a luminous efficiency of the photovoltaic device. The method for optimizing the process parameters of the photovoltaic device according to claim 3, wherein in step c, the method for measuring each of the quality characteristics is: measuring the driving voltage: providing a plurality of constant currents of the photovoltaic element, and obtaining a corresponding The operating curve of the driving voltage, or providing a plurality of constant voltages of the photoelectric element, and obtaining an operating curve corresponding to the driving current, and extracting a driving voltage under a fixed current condition; measuring the luminance of the light: using the spectroscopic light under the above fixed current condition The chromaticity measurement measures the illuminance of the specific spectrum-chroma coordinates; and the measurement of the luminous efficiency: the illuminance measured by the above and the current density corresponding to the fixed current condition are obtained by the following conversion formula: Luminous efficiency = luminous brightness / (current density x 10). According to the method for optimizing the photoelectric component process parameters of claim 1, in step d, each of the control factors and at least one higher order factor combined by at least two of the control factors are further evaluated by a statistical hypothesis. The extent of the impact. For the optimization analysis method of the photoelectric component process parameters described in claim 5, the regression equation generated in step e is Y=β0+β1X1+β2X2+...+βkXk+ε, where X1～Xk are the control of the statistical hypothesis verification. Factor or higher-order factor; β1～βk are the coefficients of each of the control factors or high-order factors. The statistical hypothesis is determined to determine whether each coefficient β1～βk is 0. If it is 0, it indicates that the control factor or high-order factor is a non-significant factor. . According to the optimization analysis method of the photoelectric component process parameters described in claim 6, the assumption of the statistical hypothesis verification is: the null hypothesis H0: β1 = β2 = ... = βk = 0 opposite hypothesis H1: β1, β2, ..., βk≠ 0 Abandonment criterion: If P>α, then reject H1 (α is the error rate, P is the probability value); the statistical decision making decision criterion is: If P<α, reject the null hypothesis, if P>>α, Then the null hypothesis is correct. For the optimization analysis method of the photoelectric component process parameters described in claim 7, in step d, the statistical hypothesis verification of the F-test is α=0.05 corresponding to rejecting the null hypothesis or the null hypothesis. The method for optimizing the process parameters of the photovoltaic device according to claim 1 is followed by the verification experiment to verify the error of each quality characteristic and the estimated value of the plurality of optimized process parameters. The method for optimizing the process parameters of the photovoltaic device according to claim 9 is characterized in that one of the quality characteristics is a large characteristic, and the optimized process parameters of each group are adapted according to the estimated value.