A high-repetition multi-pulse optical thin film life test method

Abstract

The application discloses a high-repetition multi-pulse optical thin film life test method, which finds the inflection point pulse number of the damage threshold value of a thin film element under high repetition, determines the total pulse number range of the damage threshold value test, and provides a basis for life prediction acceleration; in the range of the irradiation pulse number, the relationship between the stable value of the damage threshold value and the repetition is fitted, a thin film life prediction mechanism under high repetition is established, and the energy density allowed by stable operation of the element under other repetitions can be calculated. The application has important significance for improving the life test efficiency and accuracy of optical elements.

Description

Technical Field

[0001] This invention belongs to the field of high-power laser thin film technology, specifically relating to a method for predicting the lifetime of high-repetition-rate multi-pulse optical thin films, which is applicable to the reliability assessment and accelerated lifetime testing of optical thin films in high-power laser systems under multiple high-repetition-rate laser irradiation conditions. Background Technology

[0002] With the increasing energy and on-orbit lifespan of lasers, the optical thin-film components used must be able to withstand hundreds of millions of laser cycles. Furthermore, due to the unique operating environment of space lasers, these thin-film components are not replaceable, making long-term stability testing and lifespan prediction extremely difficult. As a weak link in laser resistance, the reliability and laser damage issues of optical thin-film components have always been one of the bottlenecks in the development of space lasers towards longer lifespans, higher power, and higher repetition rates. Therefore, revealing the lifespan prediction mechanism of optical components is crucial for maintaining the stability and safe operation of laser systems.

[0003] Existing research has shown that the damage threshold tends to stabilize after a certain number of irradiation pulses, a characteristic widely used for lifetime prediction. Currently, lifetime prediction for optical components is mainly based on the S-on-1 testing process. W. Riede of the European Space Agency conducted lifetime tests on different types of optical thin films in 2013, finding that the films exhibit a gradual stabilization trend after multiple pulse irradiations, and that laser irradiation at a certain number of pulses can be used for pre-screening of thin film samples. Furthermore, the Lawrence Livermore National Laboratory in the United States proposed using the development of damage pit size under different pulse counts to derive lifetime. This method uses high-resolution images to locate the damage boundary and contrast generated by repeated laser pulses, mapping it to the beam profile to extract the laser damage threshold and lifetime. Guo Kesheng et al. [Prior Technology CN 112326197 A] proposed predicting the relationship between the threshold and pulse count under high-repetition-rate multi-pulse laser parameters of a target by using the difference between the high and low frequency damage thresholds under a few pulse counts. Zhou Ping et al. from Dalian University of Technology [Prior Technology CN 118553322 A] proposed predicting the multi-pulse lifetime of optical thin-film components by detecting the defect content of optical thin-film components under different pulse numbers.

[0004] While current lifetime prediction methods have reduced the number of pulses required, testing efficiency remains low. Furthermore, the reliability of extrapolation methods is constrained by the assumption of consistent material damage mechanisms; if the damage mechanism changes with energy or pulse number, it may lead to lifetime prediction errors.

[0005] Lifetime testing of general optical components is typically conducted under operating conditions. Accelerated testing parameters (such as pressure and flux) are usually increased directly to expedite the test. However, the lifespan of optical components cannot be accelerated simply by increasing the pulse repetition frequency. This is because the damage performance of optical components is related to the laser repetition frequency; changing the repetition frequency may alter the damage mechanism. Therefore, reasonable accelerated testing must be comparable to the operating conditions. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing optical thin film lifetime prediction methods, such as low testing efficiency and the reliability of extrapolation methods being limited by the consistency of damage mechanisms. This invention proposes a high-repetition-rate multi-pulse accelerated testing process for optical thin film lifetime and establishes a lifetime prediction model. By finding the inflection point pulse number at which the damage threshold of the thin film element tends to stabilize at high repetition rates, and fitting the relationship between the stable damage threshold value and the repetition rate, the energy density that the element can stably operate is obtained. This is of great significance for the stability and safe operation of high-power laser systems.

[0007] The technical solution of this invention is as follows:

[0008] A method for predicting the lifetime of high-repetition-rate multi-pulse optical thin films, characterized by the following steps:

[0009] Step A. For at least one preset first test frequency F1, determine the maximum number of damage pulses N through an iterative method. S And with the number of pulses N S Based on the total number of irradiation pulses S, S-on-1 damage threshold test is performed to obtain the basic dataset of damage threshold variation with pulse number at the first test frequency F1.

[0010] Step B. For at least two second test frequencies F2 and third test frequencies F3 that are different from the first test frequency F1, repeat step A respectively to obtain the pulse number reaching a stable inflection point N. t The damage threshold stability values ​​T2 and T3 at the time are used, and combined with the damage threshold stability value T1 determined by the basic dataset at the first test frequency F1, a functional relationship T=f(F) between the repetition frequency F and the damage threshold stability value T is established.

[0011] Step C. Calculate the target frequency F based on the aforementioned functional relationship T=f(F). X At the stable inflection point N t Damage threshold stability value T X Based on the evolution of the damage threshold with the number of pulses as shown in the basic dataset, the target frequency F is constructed. X Lifetime prediction curve of damage threshold as a function of pulse number.

[0012] Furthermore, step A specifically includes:

[0013] Step A1: Set the initial maximum number of damage pulses N S , where N S ≥ 1000, and determine the total number of irradiation pulses S = 2N S ;

[0014] Step A2: Perform S-on-1 damage testing according to ISO 21254-2 standard. Monitor the damage occurrence time of each test point in real time using an online damage detection device and record the number of damage pulses M corresponding to each damage point.

[0015] Step A3: Calculate the M value of all damage points and take the maximum value M. max Determine M max ≤N S Check if the condition is true; if not, update N. S = M max And reset S = 2M max Repeat sub-step A2 until M. max ≤ N S Established;

[0016] Step A4, based on satisfying M max ≤ N S Based on the S-on-1 test results under the given conditions, calculate at least 5 different pulse numbers S. X The damage threshold below is used to form the basic dataset, where S X ≤S.

[0017] Furthermore, in step A4, the calculation of different pulse numbers S X The method for setting the damage threshold is as follows:

[0018] For any preset number of pulses S X Based on the number of damage pulses M recorded in the S-on-1 test, the original test data is reconstructed into a virtual S. X The -on-1 test dataset has the following reconstruction rule: Under any energy density E, if M ≤ S of a certain test point... X Then the test point is determined to be S. X Damage points in the -on-1 test; if M > S X If a point is identified as undamaged, it is considered an undamaged point. The number of test points identified as damaged points at each energy density E is counted, and the damage probability at that energy density is calculated. Energy density data points with damage probabilities ranging from 20% to 80% are selected for fitting, and the energy density at zero damage probability is extrapolated as S. X-on-1 damage threshold.

[0019] Furthermore, step A also includes:

[0020] Step A5: Based on the aforementioned basic dataset, plot the curve of damage threshold versus pulse number. By identifying the minimum number of pulses when the rate of change of damage threshold is less than a preset threshold, determine the stable inflection point N. t And the stable inflection point N t The corresponding damage threshold is used as the stable value of the damage threshold at this test frequency.

[0021] Furthermore, the stability inflection point N t The determination method is as follows: The relationship curve between the damage threshold and the number of pulses is fitted with an exponential decay function. When the coefficient of determination R of the fitted curve is... 2 When the percentage is greater than 95%, the damage threshold is considered to have entered a stable phase, and the number of pulses corresponding to this starting point is N. t .

[0022] Furthermore, in step B, the functional relationship T=f(F) between the repetition frequency F and the stable damage threshold T is established by linear or nonlinear regression analysis using the least squares method.

[0023] Furthermore, step C specifically includes:

[0024] Step C1: Calculate the target frequency F based on the functional relationship T=f(F). X At the stable inflection point N t Damage threshold stability value T X ;

[0025] Step C2: Obtain the target frequency F X Damage threshold data within the pulse number S ≤ 1000;

[0026] Step C3, using the damage threshold stability value T X As a long-life asymptotic value, combined with damage threshold data within the range of S ≤ 1000, the target frequency F is generated by fitting a preset lifespan model. X The life prediction curve below.

[0027] Furthermore, the preset lifetime model is an exponential decay function, the expression of which is:

[0028] LT(x) = φ + A1exp(-x / b1)

[0029] Where x is the number of irradiation pulses, LT(x) is the damage threshold for the corresponding number of pulses x, and φ is the asymptote parameter, corresponding to the stability threshold T under long lifetime. XA1 and b1 are fitting coefficients, determined from short pulse test data.

[0030] Furthermore, the high repetition rate refers to a repetition frequency F ≥ 100 Hz, and the multi-pulse refers to a cumulative pulse number S ≥ 1000.

[0031] Furthermore, through step A, N S The iterative optimization dynamically determines and minimizes the required total number of S-on-1 test pulses S, thereby accelerating the testing process; the functional relationship between the repetition frequency F and the stable value of the damage threshold T established in step B eliminates the interference of damage mechanism differences caused by changes in repetition frequency on the prediction results.

[0032] Compared with the prior art, the technical effects of the present invention are as follows:

[0033] 1) An accelerated testing procedure for optical thin film lifetime was determined by reducing the number of irradiation pulses and increasing the repetition rate;

[0034] 2) By fitting the relationship between repetition rate and damage threshold stability value to establish a lifetime curve prediction model, the energy density range of stable operation of optical thin film components under different repetition rates and pulse numbers can be theoretically calculated, which is of great significance for improving the efficiency and accuracy of optical component lifetime testing.

[0035] 3) By introducing the maximum number of damage pulses N S The iterative correction mechanism effectively reduces the total number of irradiation pulses required in S-on-1 testing, significantly improving testing efficiency. By obtaining stable damage threshold values ​​at different high repetition rates and establishing a functional relationship between frequency and damage threshold, rapid prediction of thin film lifetime at any target frequency is achieved, avoiding repeated full-band testing. Combined with the exponential decay model, a lifetime prediction curve of damage threshold as a function of pulse number is constructed, which can comprehensively reflect the performance degradation law of thin film components under long-term multi-pulse irradiation.

[0036] 4) The method of the present invention is applicable to various types of optical thin film elements, especially to the long-term reliability assessment of irreplaceable thin film elements in high-power, high-repetition-rate laser systems. Attached Figure Description

[0037] Figure 1 The transmittance spectrum of a three-wavelength antireflection thin film element;

[0038] Figure 2 This is a schematic diagram of the S-on-1 damage testing experimental setup;

[0039] Figure 3 For different S at 800Hz frequency X -on-1 damage probability curve fitting;

[0040] Figure 4 Fitting the curve of damage threshold versus pulse number at 800Hz;

[0041] Figure 5 The damage threshold varies with the number of pulses at five repetition frequencies: 500Hz, 800Hz, 1500Hz, 2500Hz, and 3300Hz.

[0042] Figure 6 The fitting relationship between the repetition frequency F and the stable value of the damage threshold is shown, and the stable value of the damage threshold at a repetition frequency of 3300 Hz is predicted.

[0043] Figure 7 Comparison of lifetime prediction curves with actual test results at a repetition rate of 3300Hz. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings, but this should not be construed as limiting the scope of protection of the present invention.

[0045] Example:

[0046] This embodiment takes the triplet antireflection film (355nmAR) commonly used in space laser system windows as an example to illustrate the specific implementation process of a method for predicting the lifetime of high repetition rate multipulse optical thin film elements.

[0047] (a) Preparation stage:

[0048] 1. First, use a spectrophotometer to test the transmittance spectrum of the sample, such as... Figure 1 As shown, its transmittance at 355 nm meets the design requirements, and the spectral curve is smooth with no obvious absorption peaks, ensuring that the initial state of the sample is consistent.

[0049] 2. Set up as follows Figure 2 The S-on-1 damage testing experimental setup is shown. A high-repetition-rate pulsed laser with an output wavelength of 355 nm and a pulse width on the nanosecond scale is used. The beam is shaped into a Gaussian distribution with a fundamental mode through a beam expander, spatial filter, and focusing lens, and then focused onto the sample surface. A He-Ne laser is used as the probe beam. When the main laser induces damage pits on the thin film surface, the scattered light signal undergoes a sudden change. Damage occurrence is detected in real time, and a shutter is triggered to cut off the main laser, while simultaneously recording the number of pulses at the time of damage occurrence.

[0050] (II) Acceleration test procedure under single frequency (taking 800Hz as an example)

[0051] 2.1. Define initial parameters:

[0052] Set the initial laser irradiation frequency F1 = 800 Hz. Set the maximum number of irradiation pulses N corresponding to damage generation at this frequency F1. S , such as NS =200000, serving as an upper limit reference for the initial exploration of the damage threshold.

[0053] 2.2. Perform S-on-1 damage threshold test and dynamically adjust the total number of pulses:

[0054] Set the upper limit of the total number of irradiation pulses to S=2N S =400000. The S-on-1 damage threshold test was conducted according to the ISO21254-2 international standard. At each test point, the sample was continuously irradiated with a laser at a repetition rate of 800 Hz, while the He-Ne scattering light detection system monitored the damage signal in real time. When a sudden increase in the scattering light signal was detected (determined as damage), irradiation at that point was immediately stopped, the number of pulses M at which the damage occurred was recorded, and the displacement platform was moved to the next test point.

[0055] After the test is completed, check the number of damaged pulses M recorded for all tests. If M > N, then... S (i.e., the damage occurs at the preset N) S (Then) explain the preset N. S If the value is too small, it will not be sufficient to cover the stable region where the damage occurred. In this case, N needs to be adjusted. S =M, and reset S=2M, and conduct supplementary tests on this energy point or related low energy points until the complete statistical regularity of damage occurrence is captured.

[0056] In this embodiment, the measured maximum number of damage pulses is less than 100,000, which satisfies the condition M ≤ 200,000, so there is no need to adjust the total number of irradiation pulses S.

[0057] 2.3. Calculate the S-on-1 damage threshold:

[0058] Count the number of damage points at each energy density level and calculate the damage probability (number of damage points / total number of test points). Ensure the damage probability covers 0-100%. Select data points with damage probabilities between 20% and 80%, and use linear regression or an error function to fit the data to obtain a damage probability curve (e.g., ...). Figure 3 (As shown). The energy density corresponding to the time when the damage probability is 0% is extrapolated to the fitted curve, and is defined as the damage threshold LIDT of the number of S-pulses at that frequency. S-on-1 .

[0059] 2.4. Derivative calculation of damage threshold (S) under different pulse numbers x -on-1)

[0060] Based on the number of damage-generated pulses M recorded in the S-on-1 damage test, S is taken as... x Calculate S (S1, S2, S3, S4…≤S, x≥5) x-on-1 damage threshold. The process is as follows: at a certain energy, if M > S in S-on-1... x Then it is believed that in S x In -on-1, no damage occurred at this point; if M≤S in S-on-1 x If the value is not specified, then that point is considered a damage point. This method is used to calculate S at different energies. x The -on-1 damage probability is used to fit the energy density at zero probability by fitting points with damage probabilities ranging from 20% to 80%, which is then used as S. x -on-1 damage threshold.

[0061] 2.5. Determining the stable inflection point and stable value

[0062] Plot the damage threshold as a function of pulse number S at this frequency (800Hz). x Changing curves (e.g.) Figure 4 (As shown in the figure). Observing the trend of the curve, it was found that as the number of pulses increases, the damage threshold gradually decreases and eventually tends to stabilize.

[0063] Determine the number of inflection point pulses N t This refers to the minimum number of pulses required when the rate of change of the damage threshold is less than a preset threshold (e.g., 1%). In this embodiment, N at 800Hz... t ≈100,000 times.

[0064] Read the stable value T1: corresponding to N t The average damage threshold at subsequent pulse counts. In this embodiment, the stability threshold T1 at 800Hz is approximately 6.78 J / cm². 2 .

[0065] 2.6. Repeatability Validation

[0066] Select at least two samples prepared in the same batch and repeat steps 2.2 to 2.5 above, taking the average value to eliminate individual differences in samples and random errors in testing.

[0067] (III) Data Acquisition and Model Building under Multiple Frequency Circulation

[0068] 3.1. Select at least two other sets of high repetition rate conditions, for example, F2=500Hz, F3=1500Hz, F4=2500Hz.

[0069] For each frequency, repeat the test procedure (steps 2.1 to 2.6) to obtain the damage threshold stability threshold for the corresponding number of pulses:

[0070] The stability threshold T2 was measured at 500 Hz.

[0071] The stability threshold T3 was measured at 1500 Hz.

[0072] The stability threshold T4 was measured at 2500 Hz.

[0073] 3.2 Establishing a correlation model between repetition frequency and stability threshold

[0074] Organize different frequencies F i and its corresponding damage threshold stability value T i Data pairs: (500, T2), (800, 6.78), (1500, T3), (2500, T4). F-values ​​are fitted using the least squares method. i With T i The relationship. Based on the data characteristics of this embodiment (such as...) Figure 5 , Figure 6 As shown in the figure, the two exhibit a strong linear negative correlation. An empirical formula is obtained through fitting:

[0075] T(F) = -0.0005 × F + 7.2153

[0076] Where T is in J / cm 2 The unit of F is Hz. This formula reflects the physical law that, under high repetition rate conditions at this stage, the thermal accumulation effect causes the thin film damage threshold to decrease linearly with increasing frequency.

[0077] (iv) Lifetime prediction and application at arbitrary repetition rates

[0078] 4.1: Predicting the stability threshold at unknown frequencies

[0079] Suppose we need to evaluate the film at F x Long-term operation capability at 3300Hz was demonstrated, but full-life testing was not conducted.

[0080] F x Substituting 3300Hz into the above fitting formula:

[0081] T pred =-0.0005×3300+7.2153=5.5653≈5.57J / cm 2 ;

[0082] 4.2: Constructing a full lifecycle prediction curve

[0083] For any given frequency F x Its stability threshold T is known. x (Obtained from step 4.1) and damage threshold data under short pulse number (S≤1000) (Obtained from step 2.4).

[0084] The complete lifetime curve at this frequency is fitted using an exponential decay function: LT(x) = φ + A1exp(-x / b1)

[0085] Where: LT(x) is the damage threshold when the number of pulses is x; x is the cumulative number of pulses; φ is the asymptote parameter, corresponding to the stability threshold T under long lifetime. x A1 and b1 are fitting coefficients, determined from short pulse test data.

[0086] Figure 7 This is a comparison chart of the lifetime prediction curve and the actual test results at a repetition rate of 3300 Hz in this embodiment. The chart shows that the prediction curve matches well with each measured data point, further verifying the applicability and accuracy of the lifetime prediction model constructed in this invention at different repetition rates.

[0087] The high repetition rate multipulse optical thin film lifetime prediction method provided by this invention significantly improves testing efficiency while ensuring prediction accuracy, and has significant engineering application value.

Claims

1. A method for testing the lifetime of high-repetition-rate multi-pulse optical thin films, characterized in that, Includes the following steps: Step A. For at least one preset first test frequency F1, determine the maximum number of damage pulses N through an iterative method. S And with the number of pulses N S Based on the total number of irradiation pulses S, S-on-1 damage threshold test is performed to obtain the basic dataset of damage threshold variation with pulse number at the first test frequency F1. Step B. For at least two second test frequencies F2 and third test frequencies F3 that are different from the first test frequency F1, repeat step A respectively to obtain the pulse number reaching a stable inflection point N. t The damage threshold stability values ​​T2 and T3 at the time are used, and combined with the damage threshold stability value T1 determined by the basic dataset at the first test frequency F1, a functional relationship T=f(F) between the repetition frequency F and the damage threshold stability value T is established. Step C. Calculate the target frequency F based on the aforementioned functional relationship T=f(F). X At the stable inflection point N t Damage threshold stability value T X Based on the evolution of the damage threshold with the number of pulses as shown in the basic dataset, the target frequency F is constructed. X Lifetime prediction curve of damage threshold as a function of pulse number.

2. The high repetition rate multi-pulse optical thin film lifetime testing method according to claim 1, characterized in that, Step A specifically includes: Step A1: Set the initial maximum number of damage pulses N S , where N S ≥ 1000, and determine the total number of irradiation pulses S = 2N S ; Step A2: Perform S-on-1 damage testing according to ISO 21254-2 standard. Monitor the damage occurrence time of each test point in real time using an online damage detection device and record the number of damage pulses M corresponding to each damage point. Step A3: Calculate the M value of all damage points and take the maximum value M. max Determine M max ≤N S Check if the condition is true; if not, update N. S = M max And reset S = 2M max Repeat sub-step A2 until M. max ≤ N S Established; Step A4, based on satisfying M max ≤ N S Based on the S-on-1 test results under the given conditions, calculate at least 5 different pulse numbers S. X The damage threshold below is used to form the basic dataset, where S X ≤S.

3. The high repetition rate multipulse optical thin film lifetime testing method according to claim 2, characterized in that, The calculation of different pulse numbers S in step A4 X The method for setting the damage threshold is as follows: For any preset number of pulses S X Based on the number of damage pulses M recorded in the S-on-1 test, the original test data is reconstructed into a virtual S. X The -on-1 test dataset has the following reconstruction rule: Under any energy density E, if M ≤ S of a certain test point... X Then the test point is determined to be S. X Damage points in the -on-1 test; if M > S X If the test point is determined to be undamaged, then the number of test points determined to be damaged at each energy density E is counted, and the damage probability at that energy density is calculated. Energy density data points with damage probabilities ranging from 20% to 80% were selected for fitting, and the energy density under zero damage probability was extrapolated as S. X -on-1 damage threshold.

4. The high repetition rate multipulse optical thin film lifetime testing method according to claim 2 or 3, characterized in that, Step A further includes: Step A5: Based on the aforementioned basic dataset, plot the curve of damage threshold versus pulse number. By identifying the minimum number of pulses when the rate of change of damage threshold is less than a preset threshold, determine the stable inflection point N. t And the stable inflection point N t The corresponding damage threshold is used as the stable value of the damage threshold at this test frequency.

5. The high repetition rate multi-pulse optical thin film lifetime testing method according to claim 4, characterized in that, The stability inflection point N t The determination method is as follows: The relationship curve between the damage threshold and the number of pulses is fitted with an exponential decay function. When the coefficient of determination R of the fitted curve is... 2 When the percentage is greater than 95%, the damage threshold is considered to have entered a stable phase, and the number of pulses corresponding to the starting point is N. t .

6. The high repetition rate multipulse optical thin film lifetime testing method according to claim 1, characterized in that, In step B, the functional relationship T=f(F) between the repetition frequency F and the stable damage threshold T is established by linear or nonlinear regression analysis using the least squares method.

7. The high repetition rate multi-pulse optical thin film lifetime testing method according to claim 1, characterized in that, Step C specifically includes: Step C1: Calculate the target frequency F based on the functional relationship T=f(F). X At the stable inflection point N t Damage threshold stability value T X ; Step C2: Obtain the target frequency F X Damage threshold data within the pulse number S ≤ 1000; Step C3, using the damage threshold stability value T X As a long-life asymptotic value, combined with damage threshold data within the range of S ≤ 1000, the target frequency F is generated by fitting a preset lifespan model. X The life prediction curve below.

8. The high repetition rate multi-pulse optical thin film lifetime testing method according to claim 7, characterized in that, The preset lifetime model is an exponential decay function, and its expression is: LT(x) = φ + A1exp(-x / b1) Where x is the number of irradiation pulses, LT(x) is the damage threshold for the corresponding number of pulses x, and φ is the asymptote parameter, corresponding to the stability threshold T under long lifetime. x A1 and b1 are fitting coefficients, determined from short pulse test data.

9. The high repetition rate multipulse optical thin film lifetime testing method according to claim 1, characterized in that, The high repetition rate refers to a repetition frequency F ≥ 100 Hz, and the multi-pulse refers to a cumulative pulse number S ≥ 1000.

10. The high repetition rate multi-pulse optical thin film lifetime testing method according to claim 1, characterized in that, Through step A, N S The iterative optimization dynamically determines and minimizes the required total number of S-on-1 test pulses S, thereby accelerating the test process; The functional relationship between the repetition frequency F and the stable damage threshold T established in step B eliminates the interference of differences in damage mechanisms caused by changes in the repetition frequency on the prediction results.

Images