Method for predicting optical film layer damage based on emission spectrum during cleaning process

Abstract

This invention discloses a method for predicting optical film damage based on emission spectra during cleaning, relating to the field of optical component damage prediction, including: S1, establishing a standard film damage database adapted to the quantitative relationship of film damage; S2, during the cleaning process, acquiring plasma emission spectra in real time and extracting the intensity of oxygen atom spectral lines. I As a spectral characteristic parameter; S3, establish the cleaning efficiency coefficient. α The invention establishes a quantitative correlation model between emission spectral characteristic parameters and damage parameters. The model includes: S4, establishing a correlation model between damage parameters and cleaning time; and S5, establishing a correlation model III between spectral characteristics and damage parameters based on correlation models I and II to predict film thickness reduction and transmittance decrease. This invention fills the technical gap in predicting surface damage during plasma cleaning by establishing a quantitative correlation model between emission spectral characteristic parameters and damage parameters of the antireflective film, significantly extending the online service life of optical components.

Description

Technical Field

[0001] This invention relates to the field of optical component damage prediction. More specifically, this invention relates to a method for predicting optical film damage based on emission spectra during the cleaning process. Background Technology

[0002] In high-power laser systems, sol-gel antireflective films on the surfaces of large-aperture optical components adsorb organic contaminants during long-term operation. This leads to a decrease in the transmittance and laser-induced damage threshold of the optical components, severely reducing the operational quality and efficiency of the laser system. Low-pressure plasma in-situ cleaning technology, due to its advantages of high efficiency, controllability, and no secondary pollution, has become the most promising in-situ cleaning method. The basic principle of plasma cleaning of organic contaminants is to degrade macromolecular organic contaminants through the physicochemical action of active groups, thereby achieving desorption from the surface of the optical film. Previous research has revealed an optimal cleaning time window for plasma cleaning. If the cleaning time is insufficient, organic contaminants are not completely removed; if the cleaning time is too long, the active groups in the energetic plasma continuously bombard the loose and porous sol-gel antireflective film, leading to the expansion of the nanopore structure on the film surface, film thinning, decreased transmittance, and irreversible damage such as a blue shift in the peak transmittance wavelength. Furthermore, a cumulative effect can occur after multiple in-situ cleanings.

[0003] Existing research has focused on evaluating the performance of plasma cleaning after the removal of organic contaminants and calculating process time. These methods provide a quantitative evaluation of plasma cleaning capabilities, but they only reflect the removal rate of organic contaminants and provide quantitative predictions of cleaning time. They do not address real-time monitoring of film damage during cleaning, nor can they predict the damage state of optical component films under given process parameters. Current plasma cleaning technologies do not correlate film damage during cleaning with plasma characteristics, making it impossible to predict the onset and extent of damage in real time. Operators can only determine whether the optical film is clean by using fixed cleaning times or offline transmittance testing, making real-time and precise damage control difficult. When plasma conditions fluctuate or the initial contamination state of components changes, fixed cleaning time parameters can easily lead to insufficient or excessive cleaning.

[0004] Therefore, there is an urgent need to develop a method that can monitor the plasma cleaning process in real time, predict the damage state of the film, and guide the determination of the cleaning endpoint, so as to achieve precise cleaning of the film on the surface of large-aperture optical components. Summary of the Invention

[0005] One object of the present invention is to solve at least the above-mentioned problems and / or defects, and to provide at least the advantages described below.

[0006] To achieve these objectives and other advantages of the present invention, a method for predicting optical film damage based on emission spectra during cleaning is provided, comprising: S1. Under the same plasma cleaning process conditions, establish a standard damage database for the film layer that is compatible with the quantitative relationship between the cleaning time t and the number of cycles m. S2. During the cleaning process, plasma emission spectra are collected in real time to extract the intensity of oxygen atom spectral lines. I As a spectral characteristic parameter; S3. Establish cleaning efficiency coefficient α Correlation Model I with spectral characteristic parameters; S4. Establish a correlation model II between damage parameters and cleaning time; S5. Based on correlation model I and correlation model II, establish correlation model III between spectral characteristics and damage parameters. Correlation model III predicts the film thickness reduction amount using the following formulas. The amount of transmission rate decrease : In the above formula, k This is the proportionality coefficient. I The intensity of the oxygen atom spectral line. n This is an empirical index obtained through experimental calibration. t opt The optimal cleaning time is given by ζ, which is the transmittance damage coefficient. η The damage-cleaning coupling coefficient; S6. During the plasma cleaning process, the emission spectrum is acquired in real time to obtain the intensity of the oxygen atom spectral lines at the current moment. I predict , and in , When the value is less than or equal to one of the preset thresholds, a stop cleaning signal is triggered.

[0007] Preferably, in S1, the process for establishing the standard damage database for the membrane layer is as follows: S10. Select a fused silica substrate prepared using the same process as large-aperture optical elements, and prepare an antireflection film on the fused silica substrate using sol-gel as a sample. S11. The sample is contaminated by fumigation so that the amount of contaminants on the sample surface is equivalent to the amount of contaminants after 24 months of operation in an actual high-power laser system. S12. The contaminated samples are cleaned in groups using a low-pressure air plasma cleaning device. The grouping refers to the following: one group is cleaned in a single cycle, with a single cleaning time of 0-30 minutes and an experiment is performed once per minute; the other group is cleaned in a cycle, with each cleaning cycle lasting 5 minutes and the number of cycles being 1-20, and each cycle is repeated 3 times. S13. Measure the transmittance of the cleaned sample at 351 nm using a spectrophotometer. T (λ), and the surface morphology and cross-sectional thickness of the film were measured using scanning electron microscopy. d ; S14. Based on the transmittance corresponding to cleaning time t and number of cycles m T、 thickness d Construct a standard damage database for membrane layers.

[0008] Preferably, in S3, the cleaning efficiency coefficient in the correlation model I is... α Characterized by Equation 1: S5. Based on correlation model I and correlation model II, establish correlation model III between spectral characteristics and damage parameters. Correlation model III predicts the film thickness reduction amount using the following formulas. The amount of transmission rate decrease : In the above formula, k This is the proportionality coefficient. I The intensity of the oxygen atom spectral line. n This is an empirical index obtained through experimental calibration. t opt The optimal cleaning time is given by ζ, which is the transmittance damage coefficient. η The damage-cleaning coupling coefficient; S6. During the plasma cleaning process, the emission spectrum is acquired in real time to obtain the intensity of the oxygen atom spectral lines at the current moment. I predict , and in , When the value is less than or equal to one of the preset thresholds, a stop cleaning signal is triggered.

[0009] In S5, the correlation model III characterizes the film thickness reduction rate using Equation 4. k d With cleaning efficiency α The relationship between them: Equation 5 characterizes the rate of decrease in transmittance. b 1. Cleaning efficiency α The relationship between them: Substituting both Equations 1 and 4 into Equation 2, we obtain the film thickness reduction. The prediction formula; Substituting both Equations 1 and 5 into Equation 3, we obtain the decrease in transmittance. The prediction formula.

[0010] Preferably, at any spectral intensity I The best time to clean t opt It is characterized by the following formula: In the above formula, I std The spectral intensity is measured under standard conditions. t std for I std The corresponding optimal plasma cleaning time.

[0011] Preferably, the process parameters for plasma cleaning are set as follows: discharge frequency of 14-16kHz, discharge pressure of 15-25Pa, and discharge voltage of 145-155V.

[0012] Preferably, the parameter set of the standard damage database for membrane layers is preferred. D ( t ) is defined as: ,in, This indicates the amount of change relative to an uncontaminated or unwashed state.

[0013] 7. The method for predicting optical film damage based on emission spectrum during the cleaning process as described in claim 1, characterized in that, in S6, The preset threshold is 5nm. The preset threshold is 0.3%.

[0014] The present invention has at least the following beneficial effects: Firstly, based on previous in-depth research on plasma cleaning mechanisms and damage patterns, this invention creatively proposes to incorporate emission spectral characteristic parameters (oxygen atom spectral line intensity) into the designation of plasma cleaning mechanisms and damage patterns. I A quantitative correlation model was established between the antireflection film and its damage parameters (film thickness reduction and transmittance decrease) to avoid insufficient or excessive cleaning during the cleaning process. This fills the technical gap in predicting material surface damage during plasma cleaning and significantly extends the online service life of optical components.

[0015] Secondly, the prediction method of the present invention is simple and reliable. It can achieve real-time monitoring and prediction by equipping a fiber optic spectrometer. There is no need to disassemble optical components for offline monitoring. The modification cost is low and it is easy to integrate. It can be widely used for in-situ maintenance of large and precision optical systems such as high-power laser devices, synchrotron radiation sources, and extreme ultraviolet lithography machines.

[0016] Third, the optical film damage prediction method of this invention during plasma cleaning can also be extended to other types of optical films and contaminants. It can be achieved simply by recalibrating the coefficients in the model according to the type of film and contaminant, and it has strong scalability.

[0017] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0018] Figure 1 This is a flowchart for predicting optical film damage during plasma cleaning. Figure 2 The evolution of film transmittance spectra under different plasma cleaning times; Figure 3 Nanopores on the film surface under different plasma cleaning times; Figure 4 The relationship between film thickness reduction and plasma cleaning; Figure 5 The graph shows the relationship between the intensity of the 844.6 nm oxygen atom spectral line in the plasma emission spectrum and the cleaning time. Figure 6 This is a graph showing the correlation between the integral value of oxygen atom spectral line intensity and the decrease in film transmittance. Figure 7 This is a comparison chart of damage monitoring and prediction results during actual plasma cleaning. Detailed Implementation

[0019] The present invention will now be described in further detail with reference to the accompanying drawings, so that those skilled in the art can implement it based on the description.

[0020] A method for predicting optical film damage during cleaning based on emission spectra is proposed. This method monitors and predicts the damage state of antireflection coatings on optical components in real time during cleaning using plasma emission spectroscopy. It is applicable to high-precision optical systems such as high-power laser devices, synchrotron radiation sources, and extreme ultraviolet lithography, where strict requirements for the cleanliness and damage control of optical components are necessary. In practice, a quantitative correlation model between plasma spectral characteristics and film damage parameters is established to achieve real-time prediction of the damage state during cleaning and accurate determination of the cleaning endpoint. The specific processing steps are as follows: Step 1: Establish a standard damage database for the film layer. Select sol-gel antireflective film samples prepared using the same process as the optical element to be cleaned, and clean them for different times under the same plasma cleaning process conditions. t A single wash cycle (0~30min), with different number of wash cycles. m (1~20 times) of cyclic cleaning (each cleaning time is...) t 0). Measure the transmittance of the antireflection membrane after cleaning. T (λ) and film thickness d Establish cleaning time. t Number of loops m The membrane damage parameters mentioned above are compared with the standard membrane damage database, wherein the parameter set in the standard membrane damage database is... D ( t ) is defined as: ,in, This indicates the amount of change relative to an uncontaminated or unwashed state.

[0021] It should be noted that in step one, the time for each cleaning cycle is set. t 0 represents 5 minutes. The plasma cleaning process parameters are: discharge frequency 14-16 kHz, discharge pressure 15-25 Pa, and discharge voltage 145-155 V. The transmittance of the antireflection membrane is measured after cleaning. T (λ) is the permeability of the antireflection membrane measured each time. T (λ) is obtained by selecting the peak value at 351 nm. Film thickness. d It is obtained by depositing an antireflective film prepared by sol-gel onto a silicon wafer to replace the fused silica substrate. After plasma cleaning, the thickness of the film is obtained by measuring the cross-sectional film thickness using SEM after the silicon wafer is cut.

[0022] Step 2: Establish spectral characteristic parameters. During the cleaning process, plasma emission spectra are acquired in real time, and the intensity I of oxygen atom spectral lines is extracted. 844.6 (Wavelength 844.6 nm) The intensity of this spectral line is proportional to the concentration of active oxygen atoms and can characterize the plasma cleaning capability.

[0023] Step 3: Establish a correlation model between the cleaning efficiency coefficient and spectral intensity. Based on the degradation mechanism of organic pollutants during plasma cleaning, define the cleaning efficiency coefficient. α This coefficient characterizes the probability of contaminants being removed per unit time; a higher value indicates higher cleaning efficiency. Experiments show that the cleaning coefficient... α With oxygen atom spectral line intensity I 844.6 The following relationship must be satisfied: In the formula,k This is the proportionality coefficient. n These are empirical indices, all of which can be obtained through experimental calibration.

[0024] Step 4: Establish a correlation model between damage parameters and cleaning time. Based on experimental data from the standard damage database for film layers, establish the correlation between film damage parameters and cleaning time, including the reduction in optical film thickness. With cleaning time t A linear relationship exists: In the formula, k d The thickness reduction rate of the optical film is expressed in nm / s.

[0025] Transmission rate The cleaning process is divided into two stages, namely the initial cleaning stage (…). t ≤ t opt As the permeability increases, organic contaminants on the membrane surface are gradually removed during this stage; in the later stages of cleaning ( t > t opt As transmittance decreases, damage gradually begins to occur on the film surface, as shown in the following formula: In the formula, t opt The optimal cleaning time is when the contaminants are just completely removed. a 1 represents the rate of increase in permeability during the pollutant removal stage. b 1 represents the rate of decrease in permeability during the membrane damage stage.

[0026] Step 5: Establish a correlation model between spectral characteristics and damage parameters. Experiments revealed that the film thickness reduction rate... k d With cleaning efficiency α Proportional: In the formula, η The damage-cleaning coupling coefficient can be calibrated experimentally. Substituting the relationship between cleaning efficiency and oxygen atom spectral line intensity from step three into the equation yields the relationship between the optical film thickness reduction and spectral intensity: Similarly, the rate of decrease in transmittance b 1 also with α Proportional: In the formula, ζ is the transmittance damage coefficient. Therefore, the formula for predicting the decrease in transmittance, based on the above formulas, is: Step Six: Prediction of the damage initiation critical point. Optimal cleaning time. t opt This can be obtained under standard conditions. The spectral intensity measured under standard conditions is set as follows: I std The optimal plasma cleaning time at this time is t std Then, any spectral intensity can be considered as I The optimal cleaning time is: Step 7: Real-time Monitoring and Damage Monitoring. During the plasma cleaning process, emission spectra are acquired in real time to obtain the current oxygen atom spectral line intensity II, and real-time monitoring and prediction are performed according to the following procedure: (1) Based on the current I Calculate the current cleaning efficiency coefficient: (2) Based on the current I Predicting the optimal cleaning time: (3) Based on the current cleaning time t Calculate the current film thickness reduction using the formula in step five. and the decrease in transmittance .

[0027] (4) When the predicted damage amount approaches the preset threshold, an alarm signal is issued and the cleaning process is automatically terminated. The preset threshold includes: film thickness reduction amount. and the decrease in transmittance , and when ≤5nm or When the concentration is ≤0.3%, a stop cleaning signal is triggered.

[0028] Step 8: Model Parameter Calibration and Update. Before using the method of this invention, it is necessary to calibrate the following parameters, namely the proportionality coefficient, through standard sample experiments. k Experience Index n Damage-cleaning coupling coefficient η Transmittance damage coefficient ζ Standard state spectral intensity I std and the corresponding optimal cleaning time t std During use, the model can be updated online based on historical cleaned data to improve prediction accuracy.

[0029] Example: The cleaning method uses fused silica-based sol-gel antireflection membranes as the target material and dibutyl phthalate (DBP) as a typical organic pollutant, so that those skilled in the art can implement the method by referring to the instructions.

[0030] (1) The process for establishing a method to predict optical film damage during plasma cleaning is as follows: Figure 1 As shown, a standard damage database needs to be established first. A fused silica substrate, prepared using the same process as for large-aperture optical elements, is selected. An antireflective coating sample is prepared on this fused silica substrate using a sol-gel method. The sample size is 50×50×5mm. 3 The sample was contaminated using a fumigation method. Specifically, a beaker containing DBP liquid was placed on a 100°C constant-temperature heating stage, with the sample positioned above the beaker. Fumigation was performed for 40 seconds, ensuring the surface contamination level was comparable to that of an actual high-power laser system after 24 months of operation. Spectrophotometer measurements showed that the transmittance of the contaminated sample at 351 nm decreased from 99.81% to 95.02%.

[0031] The contaminated samples were cleaned using a low-pressure air plasma cleaning system with discharge parameters of 20 kHz frequency, 92.5 W power, and 20 Pa pressure. The samples were divided into two groups. Group A underwent single cleaning, with cleaning times ranging from 0 to 30 minutes, conducted one experiment per minute. Each group contained three samples and was tested simultaneously. Group B underwent cyclic cleaning, with each cleaning cycle lasting 5 minutes (based on previous process parameter studies, 5 minutes is sufficient for complete contaminant removal). The number of cycles was 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 15, and 20, with each cycle repeated three times. After cleaning, the transmittance of the samples at 351 nm was measured using a spectrophotometer. T Scanning electron microscopy was used to measure the surface morphology and cross-sectional thickness of the film. d To facilitate observation of patterns, a selection of test results is shown below. Figures 2-4 As shown. (Through) Figure 2 It can be seen that under this process, the transmittance recovers to over 99.8% within a single cleaning time of less than 5 minutes. However, after 10 minutes of cleaning, the transmittance begins to gradually decrease, and the peak wavelength blue shifts. After 30 minutes of cleaning, the transmittance drops to 95.2%. Figure 4 This indicates a linear relationship between the optical film thickness reduction and the cleaning time. After 30 minutes of cleaning, the film thickness decreased by approximately 63 nm. The fitted film thickness reduction rate was [value missing]. k d =0.035nm / s.

[0032] (2) During the plasma cleaning process, a fiber optic spectrometer was used to collect the plasma emission spectrum in real time and extract the intensity of the oxygen atom spectral line at 844.6 nm. IPlace the sample in the center of the cleaning window and record the readings at different cleaning times. I value. Figure 5 Showing typical I The relationship between the values ​​and time. Multiple sets were obtained through repeated washing. α , I Data, cleaning efficiency coefficient α =2.27×10 -4 s -1 The corresponding stable phase I =111.5. An empirical index was obtained by fitting multiple sets of data. n =1.68, proportionality coefficient k = α / I n =1.52×10 -9 .

[0033] (3) Place the sample in the same position in the window for cleaning, and compare the cleaning times. I Value, such as Figure 6 As shown. Then, the film thickness reduction Δ was measured in the same way. d Calculate the position k d =Δ d / 1800. Record this position simultaneously. I Value, and by α = kI n calculate α Multiple groups ( k d , α The damage-cleaning coupling coefficient was obtained by linear fitting of the data. η =154nm. Similarly, the transmittance damage coefficient ζ is obtained by fitting, which is 0.011%·s.

[0034] (4) The established model was applied to the plasma cleaning process of actual large-aperture optical components. The component to be cleaned was 50×50×5mm. 3 The fused silica-based sol-gel antireflective membrane initially exhibited a transmittance of 99.81%, which decreased to 95.02% after contamination. The standard state of the central region of the plasma cleaning window was measured. I std =111.5, the optimal cleaning time for this area t std=300s. Place the component in the plasma cleaning equipment, and set the discharge parameters to a frequency of 20kHz, a power of 92.5W, and a pressure of 20Pa. Simultaneously with the start of cleaning, activate the fiber optic spectrometer to acquire the emission spectrum in real time, with a sampling frequency of 0.1Hz. During the cleaning process, calculate the current time in real time based on the established model. I Value, cleaning efficiency coefficient, current optimal cleaning time, current optimal cleaning time, current film thickness reduction. If t > t opt It is also necessary to calculate the current decrease in transmittance. The film thinning amount is set to 5nm and the transmittance to 0.3%. When the predicted value exceeds this threshold, the system will issue an alarm and immediately stop cleaning.

[0035] After different cleaning times, the components were removed and the film thickness reduction was measured. The trend was consistent with the model prediction, and the values ​​largely overlapped, indicating that the method can effectively predict film damage during the cleaning process. Specifically, from... Figure 7 It can be seen that the average relative error between the model predictions and the measured values ​​is 8.7%, with the largest error occurring in [the following text is incomplete and likely refers to a different topic]. t =22.8% at 12 minutes, which may be related to individual differences in samples at that time point, but the overall prediction accuracy meets the needs of engineering applications.

[0036] The above solution is merely an illustration of a preferred example and is not limited thereto. When implementing this invention, appropriate substitutions and / or modifications can be made according to the user's needs.

[0037] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. It can be applied to various fields suitable for the present invention. Other modifications can be readily made by those skilled in the art. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and examples shown and described herein.

Claims

1. A method for predicting optical film damage based on emission spectrum during cleaning, characterized in that, include: S1. Under the same plasma cleaning process conditions, establish a standard damage database for the film layer that is compatible with the quantitative relationship between the cleaning time t and the number of cycles m. S2. During the cleaning process, plasma emission spectra are collected in real time to extract the intensity of oxygen atom spectral lines. I As a spectral characteristic parameter; S3. Establish cleaning efficiency coefficient α Correlation Model I with spectral characteristic parameters; S4. Establish a correlation model II between damage parameters and cleaning time; S5. Based on correlation model I and correlation model II, establish correlation model III between spectral characteristics and damage parameters. Correlation model III predicts the film thickness reduction amount using the following formulas. The amount of transmission rate decrease : In the above formula, k This is the proportionality coefficient. I The intensity of the oxygen atom spectral line. n This is an empirical index obtained through experimental calibration. t opt The optimal cleaning time is given by ζ, which is the transmittance damage coefficient. η The damage-cleaning coupling coefficient; S6. During the plasma cleaning process, the emission spectrum is acquired in real time to obtain the intensity of the oxygen atom spectral lines at the current moment. I predict , and in , When the value is less than or equal to one of the preset thresholds, a stop cleaning signal is triggered.

2. The method of predicting damage to an optical film layer based on an emission spectrum during a cleaning process according to claim 1, wherein, In S1, the process for establishing the standard damage database for the membrane layer is as follows: S10. Select a fused silica substrate prepared using the same process as large-aperture optical elements, and prepare an antireflection film on the fused silica substrate using sol-gel as a sample. S11. The sample is contaminated by fumigation so that the amount of contaminants on the sample surface is equivalent to the amount of contaminants after 24 months of operation in an actual high-power laser system. S12. The contaminated samples are cleaned in groups using a low-pressure air plasma cleaning device. The grouping refers to the following: one group is cleaned in a single cycle, with a single cleaning time of 0-30 minutes and an experiment is performed once per minute; the other group is cleaned in a cycle, with each cleaning cycle lasting 5 minutes and the number of cycles being 1-20, and each cycle is repeated 3 times. S13. Measure the transmittance of the cleaned sample at 351 nm using a spectrophotometer T (λ) and measure the surface morphology and cross-sectional thickness of the film layer using a scanning electron microscope d ; S14, Based on cleaning time t Number of loops m Corresponding transmittance T、 thickness d Construct a standard damage database for membrane layers.

3. The method of predicting damage to an optical film layer based on an emission spectrum during a cleaning process according to claim 2, wherein, In S3, the cleaning efficiency coefficient in the correlation model I α Characterized by Equation 1: In S4, the correlation model II obtains the film thickness reduction amount through formula two based on the experimental data in the film layer standard damage database : The correlation model II is represented by Formula Three, and represents the amount of decrease in the permeability at the initial stage of cleaning and the late stage of cleaning : In the above formula, a 1 is the rate of increase in permeability in the pollutant removal stage, b 1 is the rate of decrease in permeability in the membrane layer damage stage, t opt is the optimum cleaning time; In S5, the correlation model III characterizes the film thickness thinning rate by Equation IV k d Relationship between the cleaning efficiency α and the cleaning efficiency The rate of decrease in transmittance is characterized by equation five b 1 and cleaning efficiency α relationship: Substitute formula one and formula four into formula two, to obtain the prediction formula of film thickness thinning amount . Substituting both Equations 1 and 5 into Equation 3, we obtain the decrease in transmittance. The prediction formula.

4. The method of predicting damage to an optical film layer based on an emission spectrum during a cleaning process according to claim 3, wherein, At any spectral intensity I The best time to clean t opt It is characterized by the following formula: In the above formula, I std The spectral intensity is measured under standard conditions. t std for I std The corresponding optimal plasma cleaning time.

5. The method of predicting damage to an optical film layer based on an emission spectrum during a cleaning process according to claim 2, wherein, The process parameters for plasma cleaning are set as follows: discharge frequency of 14-16kHz, discharge pressure of 15-25Pa, and discharge voltage of 145-155V.

6. The method of predicting damage to an optical film layer based on an emission spectrum during a cleaning process according to claim 2, wherein, Parameter set of film layer standard damage database D ( t ) is defined as: wherein, denotes the change amount with respect to the unsoiled, unwashed state.

7. The method of predicting damage to an optical film layer based on an emission spectrum during a cleaning process of claim 1, wherein, In S6, The preset threshold is 5 nm, The preset threshold is 0.3%.

Images