Large-size substrate ink-jet printing film packaging online quality detection method and system

Abstract

The invention belongs to the related technical field of ink-jet printing manufacturing, and particularly relates to a large-size substrate ink-jet printing film packaging online quality detection method and system, and the method comprises the steps: employing a spectrum sensor to measure the thickness of a plurality of measurement points of a current to-be-detected film online, and correcting a value which deviates from the average film thickness level greatly; obtaining a dispersion equation corresponding to the current to-be-measured film from a pre-constructed database, and determining the wavelength of light irradiated to each measuring point on the upper surface of the film by the sensor; obtaining the film refractive index of the corresponding measuring point according to the optical wavelength and the dispersion equation; calculating the optical path difference of the sensor light beam in the film based on the film thickness sample value of each measuring point, and correcting the true value of the film at the corresponding measuring point based on the relationship between the optical path difference and the refractive index of the film; and adopting a spectrum confocal sensor to measure and correct the heights of multiple measuring points in a preset area of the edge of the current to-be-measured film on line to obtain the real shape of the edge. The invention relates to an online high-efficiency quality detection method for ink-jet printing film packaging of a large-size substrate.

Description

Technical Field

[0001] The present invention belongs to the technical field related to inkjet printing manufacturing, and more specifically, relates to an on-line quality detection method and system for inkjet printing film encapsulation on a large-size substrate. Background Art

[0002] With the development of inkjet printing technology, the development of inkjet printing equipment for larger-area film encapsulation has become an important trend. One of the key links in the manufacturing of large-area film encapsulation by inkjet printing is the large-area inkjet printing film quality detection technology, which can detect the thickness uniformity and edge morphology of the printed film after printing. Through on-line detection, unqualified films can be found and excluded in time, thereby reducing losses and rework in subsequent production; detecting during the production process can help adjust printing parameters (such as temperature, pressure, materials, etc.), thereby further optimizing the film preparation process and making it more stable and efficient.

[0003] The detection of large-area inkjet printing films has introduced new challenges and problems. A patent CN201711183118.2 describes a large-area dynamic measurement device and method for the preparation process of nano films, but the so-called large area here does not meet the current measurement requirements for meter-level large areas, and it cannot handle the measurement of film thickness above 10 microns, nor can it meet the construction of film morphology, and no measurement method for G6-size and higher-generation-size inkjet large-area films is proposed. Patent CN202210083938.9 describes a rapid large-area film quality detection method, but it covers the large-area film to be detected with color-developing ink, which damages the manufactured film itself, cannot effectively save costs, and cannot accurately obtain the edge morphology of the film.

[0004] Further research finds that the existing measurement technologies in patents and literature still have the following deficiencies:

[0005] 1. Under the conditions of large-area film detection, the deformation of the film surface height caused by the conveying and loading methods and the film thickness measurement fluctuations caused by the uneven distribution of the incident wavelength of the optical system have not been effectively processed.

[0006] 2. For the thickness measurement and morphology measurement of the film edge region with an inclined morphology, the existing point interference measurement method cannot return the thickness of the edge region, and the white light interference method is difficult to efficiently obtain the edge morphology.

[0007] 3. For the detection of transparent films, the signal reflection between the film and the substrate is weak, and the parameters of the film thickness cannot be effectively extracted. Especially under the principle of spectral confocal, the reflection signal is even weaker, and it is difficult to directly obtain the film thickness. Summary of the Invention

[0008] In view of the above defects or improvement requirements of the prior art, the present invention provides an online quality detection method and system for inkjet printing film encapsulation of large-size substrates, aiming to propose an online high-efficiency quality detection method for inkjet printing film encapsulation of large-size substrates.

[0009] To achieve the above object, according to one aspect of the present invention, there is provided an online quality detection method for inkjet printing film encapsulation of large-size substrates, including:

[0010] Using a spectral sensor to online measure the thickness of multiple measurement points of the current film to be measured, forming a film thickness sample set, calculating the Z-score of the film thickness of each measurement point, and when it is greater than the threshold, using the average film thickness of two adjacent measurement points in the surface space of the film at this measurement point to replace the film thickness data of this measurement point in the film thickness sample set, forming a new film thickness sample set;

[0011] Obtaining the dispersion equation corresponding to the current film to be measured from a pre-constructed database, and determining the light wavelengths at each measurement point on the upper surface of the film irradiated by the sensor; obtaining the film refractive index at the corresponding measurement point from the light wavelength and the dispersion equation; calculating the optical path difference of the sensor beam in the film based on the film thickness sample values of each measurement point in the new film thickness sample set, and based on the relationship between the optical path difference and the film refractive index, combined with the film refractive index at the corresponding measurement point, correcting to obtain the true value of the film at the corresponding measurement point to quantify the film thickness uniformity of the current film to be measured;

[0012] Using a spectral confocal sensor to online measure and correct the height of multiple measurement points at the edge of the current film to be measured to obtain the true edge topography; wherein, the multiple measurement points at the edge are located within four preset effective edge measurement regions, and the upper left coordinates are respectively (a 1 -n, b 1 +n), (a 2 -s, b 1 +n), (a 1 +s, b 2 +s), (a 1 +s, b 1 +n), and the lower right coordinates are respectively (a 1 +s, b 2 -n), (a 2 +n, b 2 -n), (a 2 -s, b 2 -n), (a 2 -s, b 1 -s); in the formula, (a 1 , b 1 ), (a 2 , b 2are the coordinates of the upper left corner and the lower right corner of the printing area determined based on the printed vector graphic element information respectively. s = α * (m % d), n = d - s. s represents the maximum length occupied by the film edge in the height measurement field of view where the film first appears. m represents the estimated maximum climbing distance of the film. d represents the line width of the single measurement field of view of the spectral confocal sensor. α is a preset proportionality coefficient. d represents the length occupied by the substrate in the corresponding height measurement field of view.

[0013] Further, when measuring the thickness of multiple measurement points in the central effective area of the film to be measured using a line spectral interference sensor, the measurement point area is the central effective rectangular area determined based on the known printed graphic element vector data information, expressed as:

[0014]

[0015] In the formula, (a 1 , b 1 ), (a 2 , b 2 ) are the coordinates of the upper left corner and the lower right corner of the printing area determined based on the printed vector graphic element information respectively. (a 1 ′, b 1 ′), (a 2 ′, b 2 ′) are the coordinates of the upper left corner and the lower right corner of the central effective rectangular area respectively. n a is the integer value obtained by taking the square root of the total number of preset measurement points. n v is the integer value obtained by taking the square root of the known number of effective measurement points.

[0016] Further, when measuring the thickness of multiple measurement points in the four edge effective measurement areas of the current film to be measured using a spectral confocal sensor, it further includes: performing a fusion process on the film thickness of the measurement points where the central effective rectangular area and the four edge effective measurement areas overlap, and taking the combined thickness measurement value after the fusion process as the film thickness measurement value corresponding to the measurement point for Z-score calculation. Among them, the fusion process method is:

[0017]

[0018] Among them, t ic is the combined thickness measurement value, t c is the thickness measurement value of the spectral confocal sensor for the overlapping measurement points, t i is the thickness measurement value of the spectral sensor for the overlapping measurement points, σ c is the standard deviation of the thickness measurement value of the spectral confocal sensor, σ i is the standard deviation of the thickness measurement value of the spectral sensor.

[0019] Further, the thin film to be measured includes a liquid film and a cured film, corresponding to a new liquid film thickness sample set and a new cured film thickness sample set respectively;

[0020] Among them, the estimated value of the printed liquid film thickness is calculated using the printing input parameters and used as the theoretical mean value of the liquid film thickness for the Z-score calculation of the liquid film thickness samples; the known target printed film thickness value is used as the theoretical mean value of the cured film thickness for the Z-score calculation of the cured film thickness samples;

[0021] The Z-score of the liquid film thickness at each measurement point is:

[0022]

[0023] In the formula, is the Z-score value of the i-th measurement point of the liquid film, is the measured value of the liquid film thickness at the i-th measurement point of the liquid film, is the mean value of the liquid film thickness of all measurement points, m is the number of liquid film thickness measurement points; is the estimated value of the printed liquid film thickness, n p 、n e are the number of printed pixels and the number of thin film edge pixels extracted determined based on the known printed primitive vector data information respectively, is the known average printed droplet volume, S is the printed thin film area determined based on the edge information of the known printed primitive vector data;

[0024] The Z-score of the cured film thickness at each measurement point is:

[0025]

[0026] In the formula, is the Z-score of the i-th measurement point of the cured film, is the measured value of the cured film thickness at the i-th measurement point of the cured film, is the theoretical mean value of the cured film thickness, is the mean value of the cured film thickness of all measurement points, m is the number of cured film thickness measurement points of all measurement points.

[0027] Further, the construction method of the dispersion equation corresponding to various thin films in the database is:

[0028] Regarding the stacked structure of the target thin film and the target substrate as a double-layer measurement medium structure, under the detection light of different wavelengths λ, the distance x 1 ′ from the sensor head to the upper surface of the double-layer measurement medium structure, the distance x 2 ′ from the sensor head to the lower surface of the double-layer measurement medium structure, and the distance x from the sensor head to the upper surface of the mirror3 '; Based on n b (λ) and the target substrate thickness H, combined with the geometric relationship of beam propagation, calculate the refractive index of the target thin film for light of each wavelength to fit and obtain the dispersion equation n f (λ), where n b (λ) represents the dispersion equation of the target substrate medium;

[0029] Calculate the refractive index of the target thin film for light of each wavelength through the following formula:

[0030]

[0031] In the formula, n f，λ is the refractive index of the target thin film for light with a wavelength of λ, θ 2 is the refraction angle of light propagating in the thin film layer, θ 3 is the refraction angle of light propagating in the substrate.

[0032] Furthermore, when using a spectral interference sensor to detect the film thickness at multiple measurement points of the thin film to be measured located on the substrate, mount a spectral confocal sensor on the moving axis of the spectral interference sensor to synchronously measure the height information at multiple points on the substrate, and perform the following correction on the measured film thickness samples:

[0033] Based on the height information at multiple points on the substrate, combined with the air bearing system parameters, reconstruct the height of the upper surface of the substrate;

[0034] According to the reconstructed height information of the upper surface of the substrate and the dispersion equation of the thin film to be measured, correct the film thickness at each measurement point of the thin film to be measured. The correction method is:

[0035] t f (x,y) = s f (x,y) / (n f (λ(h f (x, y)))

[0036] In the formula, s f (x, y) represents the measured value of the sensor head for the film thickness at the coordinates (x, y) on the upper surface of the thin film, h f (x, y) represents the height at the coordinates (x, y) on the upper surface of the thin film, λ(h f (x, y)) represents the light wavelength irradiated at the coordinates (x, y) on the upper surface of the thin film; n f (λ(h f (x, y))) represents the refractive index at the coordinates (x, y) on the upper surface of the thin film.

[0037] Furthermore, when using a spectral confocal sensor to detect the film thickness at multiple measurement points of the current thin film to be measured located on the substrate, the correction method for the measured film thickness samples is:

[0038] Based on the current substrate thickness, refractive index, and the refractive index of the current thin film (the index is the thin film material and type), correct the thickness of the current substrate to obtain the actual thickness of the current substrate;

[0039] Based on the actual substrate thickness, correct the measured film thickness at the film measurement point:

[0040]

[0041] Among them, t f (x, y) is the actual thickness at the thin film coordinate (x, y), α f (x, y) is the incident angle of the measurement beam on the thin film, s f (x, y) is the measured thickness at the thin film coordinate x, λ(h f (x, y)) is the light wavelength corresponding to the measured height at the upper surface coordinate (x, y) of the thin film; n f is the refractive index of the thin film, n b is the refractive index of the substrate, n f and n b The values of and n f (x, y)) are determined based on λ(h b (x, y) is the actual thickness at the current substrate coordinate (x, y), and its value is obtained by correcting the current substrate thickness based on the current substrate thickness, refractive index, and the refractive index of the current thin film.

[0042] Furthermore, the method for quantifying the film thickness uniformity of the current thin film to be measured is:

[0043]

[0044] In the formula, U is the quantified value of the film thickness uniformity. If it is greater than the target threshold, the film thickness meets the uniformity index requirements, t f is the new film thickness sample set of the current thin film to be measured.

[0045] Furthermore, it also includes: According to the ramp height or ramp distance at the edge of each effective measurement area on the edge, measure whether the edge topography meets the production requirements, and the measurement standard calculation formula is:

[0046]

[0047] Among them, ΔH is the edge topography height evaluation index, Δh is the ramp height of the thin film edge, h f is the average value of the film thickness samples in the new liquid film thickness sample set or the new cured film thickness sample set.

[0048] According to another aspect of the present invention, there is provided an on-line quality inspection system for inkjet printing film encapsulation of large-size substrates, which is used to execute an on-line quality inspection method for inkjet printing film encapsulation of large-size substrates as described above, including: an on-line quality inspection module and a main control processor;

[0049] The on-line quality inspection module includes a film internal thickness distribution measurement sub-module and a printing edge morphology detection sub-module;

[0050] The main control processor is used to control the film internal thickness distribution measurement sub-module to measure the film thickness through a spectral sensor, control the printing edge morphology detection sub-module to measure the edge morphology through a spectral confocal sensor, and perform film thickness correction, film thickness uniformity characterization, edge morphology correction and edge morphology evaluation processing based on the measurement data.

[0051] Generally speaking, compared with the prior art by the above technical solutions conceived by the present invention, the technical solutions provided by the present invention mainly have the following beneficial effects:

[0052] 1. The present invention proposes an on-line quality inspection method for inkjet printing film encapsulation of large-size substrates. By using the Z-score evaluation method, abnormal fluctuations in measurement values caused by dust or sudden changes in the optical properties of the medium during the measurement process are eliminated, and information on film morphology changes and thickness changes caused by phenomena such as particles and mura is retained. Considering that under large-area substrates, the deformation of the substrate surface is affected by the loading method and transportation method, resulting in differences in the optical path difference of the light incident on the film to be measured by the optical instrument, a spectral sensor with the distribution of optical wavelength dispersion along the measurement height is selected. Without performing Z-direction scanning, the distribution of optical wavelengths and the incident angles at different heights are obtained to correct the actual thickness measurement error caused by changes in the measurement surface height and incident angle, which can avoid the defocus phenomenon of the small-range measurement system for measuring large-area films, and at the same time reduce the time consumption in the method of first constructing the entire surface height and then dynamically tracking the surface height for measurement, meeting the requirements for measuring large-area films. The spectral confocal method is adopted to measure the edge morphology. Based on the calculation of the measurement field of view size and the edge morphology length, the number and area of the selected edge regions are minimized, reducing the scanning time consumption, which is beneficial to improving the measurement efficiency and obtaining a complete edge at high speed, reducing the measurement result error caused by the vibration of the film carrier substrate, and meeting the requirements for on-line large-area measurement of the film edge morphology.

[0053] 2. The present invention provides a method for extracting the effective measurement region of spectral interference. By using the vector map information of the thin film to be measured and the number of effective measurement points, the central effective rectangular region is extracted. Firstly, it avoids the influence of retaining invalid measurement points in the angularly inclined region on the final uniformity evaluation, improving the data quality of the sample set containing film thickness data. Secondly, for large-area thin films, it can effectively reduce the size of the measurement domain, shorten the measurement time, and improve the efficiency of obtaining the sample set. It can also adapt to the extraction of the central effective region of thin films with peculiar non-rectangular shapes. Extracting the effective rectangular region facilitates splicing and fusion with the edge region to obtain a complete topography, avoiding the measurement method and fusion difficulties caused by special shapes.

[0054] 3. The present invention provides a method for fusing the film thickness of the measurement points where the central effective rectangular region coincides with the four edge effective measurement regions. The overlapping region between the central region measured by the spectral interference sensor and the edge region obtained by the spectral confocal sensor is fused. Firstly, it can improve the data quality of the measurement results in the overlapping region. The film thickness measured by the spectral interference is affected by the angle of the measurement surface, and the numerical value of its result is relatively accurate, but the change trend is unstable. The numerical accuracy of the film thickness measured by the spectral confocal is not as good as that of the spectral interference, but the change trend brought by the topography information is relatively stable. By fusing the measurement results of the two, the thickness information of the overlapping region can be obtained with higher accuracy.

[0055] 4. The present invention provides a method for calibrating the film thickness measured by a spectral interference sensor. The instrument needs to disperse the optical wavelength in space, and then calculate the true film thickness according to the optical path difference obtained at different heights and the film dispersion equation, avoiding the inaccuracy of the optical wavelength value incident on the surface of the thin film to be measured caused by different spectral widths of the spectral interference sensor system, resulting in uneven calculation of the film thickness. This method can effectively solve the problem of thickness measurement error of thin films with strong dispersion under the influence of large-area height distribution. For thin films with weak dispersion, it can improve the accuracy, which is superior to the refractive index averaging method and can handle more diverse and larger-area thin film measurement scenarios.

[0056] 5. The present invention provides a method for calibrating the film thickness measured by a spectral confocal sensor. By dispersing the optical wavelength in space and combining the surface height, substrate thickness, substrate dispersion equation, and film dispersion equation, the true film thickness is calculated. It can extract sub-micron film thickness when the spectral confocal sensor has a wide range, and at the same time, it can also reduce the calculation error of the true thickness caused by the change of film height under a large area. It can realize the extraction of thinner film thickness and the extraction of film thickness with low interface refractive index difference, and is applicable to more types of film thickness measurement scenarios. Description of the Drawings

[0057] Figure 1It is a flowchart of an on-line quality detection method for inkjet printing film encapsulation of large-size substrates provided by an embodiment of the present invention;

[0058] Figure 2 It is a schematic diagram of the measurement area and measurement scanning path of the film under the sensor group provided by an embodiment of the present invention;

[0059] Figure 3 It is a schematic flowchart of a method for obtaining film layer thickness by combining spectral confocal and spectral interference provided by an embodiment of the present invention;

[0060] Figure 4 It is a schematic flowchart of the film full-surface scanning provided by an embodiment of the present invention;

[0061] Figure 5 It is a schematic flowchart of a thickness measurement correction method when the refractive indices of the two layers of the spectral confocal bilayer film are similar provided by an embodiment of the present invention;

[0062] Figure 6 It is a schematic diagram of an on-line quality detection system for inkjet printing film encapsulation of large-size substrates provided by an embodiment of the present invention.

[0063] In all the drawings, the same reference numerals are used to represent the same elements or structures, where:

[0064] 11 is the first motion component, 12 is the first motion control component, 13 is the air-bearing platform, 14 is the first pressure control component, 21 is the point spectral confocal measurement head, 22 is the first control and analysis component, 23 is the second motion component, 24 is the line spectral interference measurement head, 25 is the second control and analysis component, 26 is the second motion control component, 31 is the third motion component, 32 is the line spectral confocal measurement head, 33 is the fourth motion component, 34 is the third motion control component, 35 is the third control and analysis component, 36 is the fourth motion control component, 41 is the film to be measured, 42 is the substrate to be measured, and 5 is the main control processor. Detailed implementation manners

[0065] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0066] Embodiment 1

[0067] An on-line quality detection method for inkjet printing film encapsulation of large-size substrates, as Figure 1 shown, includes:

[0068] An online measurement is carried out on the thickness of multiple measurement points of the currently measured thin film by using a spectral sensor to form a film thickness sample set. The Z-score of the film thickness of each measurement point is calculated. When it is greater than the threshold, the average value of the film thickness of two measurement points adjacent to the current measurement point in the surface space of the thin film is used to replace the film thickness data of this measurement point in the film thickness sample set to form a new film thickness sample set;

[0069] The dispersion equation corresponding to the currently measured thin film is obtained from a pre-constructed database, and the optical wavelength at each measurement point on the upper surface of the thin film irradiated by the sensor is determined; the refractive index of the thin film at the corresponding measurement point is obtained from the optical wavelength and the dispersion equation; the optical path difference of the sensor beam in the thin film is calculated based on the film thickness sample values of each measurement point in the new film thickness sample set. Based on the relationship between the optical path difference and the refractive index of the thin film, combined with the refractive index of the thin film at the corresponding measurement point, the true value of the thin film at the corresponding measurement point is corrected to quantify the film thickness uniformity of the currently measured thin film;

[0070] An online measurement and correction are carried out on the heights of multiple measurement points at the edge of the currently measured thin film by using a spectral confocal sensor to obtain the true edge topography; among them, the multiple measurement points at the edge are located within four preset effective edge measurement regions, and the upper left coordinates are respectively, (a 2 -s, b 1 +n), (a 1 +s, b 2 +s), (a 1 +s, b 1 +n), and the lower right coordinates are respectively (a 1 +s, b 2 -n), (a 2 +n, b 2 -n), (a 2 -s, b 2 -n), (a 2 -s, b 1 -s); in the formula, (a 1 , b 1 ), (a 2 , b 2 ) are respectively the upper left and lower right coordinates of the printing area determined based on the printing vector graphic element information, s = α*(m%d), nn = d - s, s represents the maximum length occupied by the thin film edge in the height measurement field of view where the thin film first appears, m represents the estimated maximum climbing distance of the thin film, d represents the line width of the single measurement field of view of the spectral confocal sensor, α is a preset proportional coefficient, and d represents the length occupied by the substrate in the corresponding height measurement field of view.

[0071] As a preferred embodiment, when measuring the thickness of multiple measurement points in the central effective area of the thin film to be measured using a line spectrum interference sensor, the starting position of the measurement in the central effective area is determined by the input printed vector graphic element information, and the measurement area of the printed thin film in the central effective area is selected as a rectangle. Considering that there may be inclined microtopographies at the edges of the printed thin film, the sensor cannot obtain measurement values at such positions under the principle of spectral interference measurement, and there are missing values in the measurement results. Therefore, the following method is used to select the measurement domain:

[0072]

[0073]

[0074] In the formula, (a 1 , b 1 ), (a 2 , b 2 ) are the upper left and lower right coordinates of the printed domain determined based on the printed vector graphic element information respectively, (a 1 ′, b 1 ′), (a′ 2 , b′ 2 ) are the upper left and lower right coordinates of the central effective rectangular area respectively, n a is the integer value obtained by taking the square root of the total number of preset measurement points, and n v is the integer value obtained by taking the square root of the known number of effective side measurement points.

[0075] As a preferred embodiment, when measuring the thickness of multiple measurement points in the four edge effective measurement areas of the current thin film to be measured using a spectral confocal sensor, it further includes: performing a fusion process on the film thickness of the measurement points where the above central effective rectangular area and the four edge effective measurement areas overlap, and using the comprehensive thickness measurement value after the fusion process as the film thickness measurement value corresponding to the measurement points for Z-score calculation. Among them, the fusion process method is:

[0076]

[0077] Among them, t ic is the comprehensive thickness measurement value, t c is the thickness measurement value of the spectral confocal sensor for the overlapping measurement points, t i is the thickness measurement value of the spectral sensor for the overlapping measurement points, σ c is the standard deviation of the thickness measurement value of the spectral confocal sensor, and σ i is the standard deviation of the thickness measurement value of the spectral sensor.

[0078] After completing the step-by-step measurements of each sensor, the following can be obtained as Figure 2For the thin film measurement area distribution map shown, this case proposes a multi-sensor data fusion and error compensation method. Regarding the multi-sensor data fusion and error compensation method of this preferred method: The thickness distribution of the central area of the thin film is obtained by a spectral sensor, and the edge topography information scanned by a spectral confocal sensor is obtained, and it is fused with the average thickness information of the overlapping scanning area of the two sensors to construct a distribution map of the thin film thickness. Combining the substrate topography obtained by the point spectral confocal sensor, the topography of the thin film on the entire substrate is completely reconstructed to form a complete construction of the large-area thin film topography and thickness.

[0079] Among them, the size of the overlapping scanning area of the two sensors is determined by the spectral confocal scanning width d and the coordinates of the effective measurement domain of spectral interference (a 1 ′, b 1 ′), (a 2 ′, b 2 ′). Within this range, the measurement result of the thickness is obtained by fusing the measurement results of the two sensors:

[0080]

[0081] Among them, t ic is the comprehensive thickness measurement value, t c is the confocal thickness measurement value, t i is the spectral interference thickness measurement value, σ c is the standard deviation of the confocal measurement value, σ i is the standard deviation of the spectral interference measurement value.

[0082] To evaluate the accuracy of thickness compensation in the two measurements, use the measurement results of a step profiler or FIB to calculate the thickness compensation accuracy ε = t r / t g , where ε is the accuracy evaluation value, t r is the true measurement result obtained by the step profiler or FIB, t g is the measurement value obtained after calibration. In the confocal measurement domain, it is t c , in the spectral interference measurement domain, it is t i , and in the common measurement domain, it is t ic . If 1 - ε ≥ ±5% in a certain domain, it does not meet the requirements, and the dispersion curve needs to be refitted or the sensor measurement coefficient needs to be corrected.

[0083] Specifically, the process of correcting the edge measurement information for the fused sensor measurement data is as Figure 3 shown. Combining the air bearing parameters with the surface height values detected by the point and line confocal sensors to predict the topography and thickness of the substrate. In Figure 2In the overlapping fusion area, the measured value of the spectral interference sensor is used as verification data to compare with the corrected value of the edge film thickness measurement. After meeting the requirements, the measured values of the edge film thickness and topography are weighted and fused according to the measurement accuracy difference between spectral interference and spectral confocal to obtain the final edge measurement value, which is finally combined with the true central thickness distribution to complete the measurement of the entire film surface.

[0084] Finally, by combining the liquid film thickness distribution and topography information before curing, the curing parameter information, and the cured film thickness distribution and topography information, analyze the influence of the internal parameters of the curing chamber on the topography of the final cured film during the curing process, guide the regulation of parameters such as the air pressure, temperature, and light intensity in the curing chamber, detect the indicators of the curing system, eliminate abnormal conditions in the curing chamber, and improve production stability.

[0085] Establish databases for substrate properties, air bearing system properties, solution properties, printing process parameters, nozzle hole properties, curing chamber parameters, and cured film parameters to provide reference for the process adjustment of subsequent printing processes.

[0086] As a preferred implementation mode, the thin film to be measured includes a liquid film and a cured film, corresponding to a new liquid film thickness sample set and a new cured film thickness sample set respectively; among them, the estimated value of the printed liquid film thickness is calculated using the printing input parameters as the theoretical mean of the liquid film thickness for Z-score calculation of the liquid film thickness sample, and the standard deviation of all samples in the liquid film thickness sample set is used as the standard deviation of the measured result of the liquid film thickness for Z-score calculation; the known target printed film thickness value is used as the theoretical mean of the cured film thickness for Z-score calculation of the cured film thickness sample.

[0087] The Z-score of the liquid film thickness for each measurement point:

[0088] According to the printing primitive vector data information required during the printing process (including the number of printed patterned sampling points) and the average printed droplet volume, calculate the total volume of the liquid film, that is According to the edge information of the input vector data, obtain the area of the printed thin film; pre-evaluate the thickness information of the printed liquid film, correct the liquid film dispersion problem, and correct the abnormal measured values of the cured film and the liquid film parts.

[0089] During the printing process, the estimated value of the liquid film thickness of the liquid film is calculated using the printing input parameters, and the calculation formula is as follows:

[0090]

[0091] Among them, is the estimated liquid film thickness, n p is the number of pixel points, n e is the number of extracted edge pixels, is the average printed droplet volume, and S is the area of the printed thin film.

[0092] Use As the theoretical mean of the liquid film thickness measurement results, the liquid film thickness at different positions of the liquid film is measured using a spectral interference sensor to form a liquid film thickness sample set, and the standard deviation of the liquid film thickness at all measurement positions is calculated. Calculate the Z-score of the liquid film thickness at each measurement position:

[0093]

[0094] Wherein, Is the Z-score of the i-th measurement position point of the liquid film. When it is greater than the threshold, the liquid film thickness data at this measurement position point is deleted (filtered) and the average value of the liquid film thickness at two measurement position points adjacent to this measurement position point in the liquid film surface space is used as compensation (using the average value of the liquid film thickness at two measurement position points adjacent to this measurement position point in the liquid film surface space to replace the liquid film thickness data at this measurement position point in the liquid film thickness sample set to form a new liquid film thickness sample set), Is the measurement value of the i-th measurement point of the liquid film, Is the calculated theoretical mean of the liquid film thickness, Is the average value of the liquid film thickness at all measurement positions, and m is the number of liquid film thickness measurement position points.

[0095] Regarding the Z-score of the cured film thickness at each measurement point:

[0096] The cured film is selected to use the target film thickness value as the theoretical mean (known quantity) to evaluate the measured cured film, and calculate the Z-score of each measurement data:

[0097]

[0098] Wherein, Is the Z-score of the i-th measurement point of the cured film. When it is greater than the threshold, this measurement result is filtered and compensated with the average value of the previous and next two measurement points, Is the measurement value of the i-th measurement point of the cured film, Is the theoretical mean of the cured film thickness, Is the average value of the cured film thickness measured at different positions of the cured film using a spectral interference sensor, and m is the number of cured film thickness measurement points.

[0099] As a preferred implementation manner, the construction method of the dispersion equation corresponding to various thin films in the above database is:

[0100] Under the detection light of different wavelengths λ, the distances from the sensor head to the upper surface of the single-layer measurement medium, the distances from the sensor head to the lower surface of the single-layer measurement medium, and the distances from the sensor head to the upper surface of the mirror are measured offline; based on the geometric relationship of beam propagation and the law of refraction, the refractive indices of the single-layer target substrate medium for lights of various wavelengths are calculated to fit the dispersion equation n b (λ); where the mirror is located directly below the medium;

[0101] The refractive index of the single-layer target substrate medium for the detection light with wavelength λ is calculated by the following formula:

[0102]

[0103] In the formula, n b,λ is the refractive index of the single-layer target substrate medium for light with wavelength λ; x 1 , x 2 , x 3 are the distances from the sensor head to the upper surface of the single-layer measurement medium, the distances from the sensor head to the lower surface of the single-layer measurement medium, and the distances from the sensor head to the upper surface of the mirror obtained by the sensor during the calibration process, and r len is the radius value of the sensor dispersion lens;

[0104] Taking the laminated structure of the target thin film and the target substrate as a double-layer measurement medium structure, under the detection light of different wavelengths λ, the distances x 1 ′ from the sensor head to the upper surface of the double-layer measurement medium structure, the distances x 2 ′ from the sensor head to the lower surface of the double-layer measurement medium structure, and the distances x 3 ′ from the sensor head to the upper surface of the mirror are measured offline; based on n b (λ) and the target substrate thickness H, combined with the geometric relationship of beam propagation, the refractive indices of the target thin film for lights of various wavelengths are calculated to fit the dispersion equation n f (λ) of the target thin film, where n b (λ) represents the dispersion equation of the target substrate medium;

[0105] The refractive indices of the target thin film for lights of various wavelengths are calculated by the following formula:

[0106]

[0107] In the formula, n f，λ is the refractive index of the target thin film for light with wavelength λ, θ 2 is the refraction angle of light propagating in the thin film layer, and θ 3 is the refraction angle of light propagating in the substrate.

[0108] Specifically, in the calibration before inkjet printing, the existing methods generally use ellipsometers for thin film thickness measurement and thin film refractive index calibration. However, due to the high cost of ellipsometers and the difficulty of integrating them into the system for in-machine measurement, it is necessary to adopt spectral interference and spectral confocal sensors that are more suitable for online high-efficiency measurement. For thin film encapsulation, since its refractive index is close to that of the glass substrate, spectral interference or confocal sensors will inevitably encounter the problem of weak interface optical signals. This results in the inability of point-type or line-type spectral interference and confocal sensors to achieve the dielectric separation between the thin film and the glass substrate, thus introducing the need for optical parameter calibration of both double layers of media during offline operation and also posing requirements for the optical uniformity of the media. Based on this background, this case proposes a method for calibrating double-layer media with a spectral confocal sensor. The specific offline calibration process is as follows:

[0109] Use light of multiple different wavelengths in spectral confocal to calibrate materials of different types, different thicknesses, and different concentrations (liquid), and fit the dispersion equations under different properties. In this scenario, the measured media are all transparent media, and the Cauchy dispersion equation is used for fitting. The equation is as follows:

[0110]

[0111] where λ is the wavelength of the light focused on the surface of the measured medium, a, b, and c are fitting coefficients adapted to materials with different properties, and n(λ) is the refractive index of the light with wavelength λ passing through this medium.

[0112] The optical sensor calibrates the optical parameters (i.e., refractive index) of the substrate and the thin film respectively, that is, obtains the refractive indices of the substrate, liquid printing solution, and cured thin film before subsequent manufacturing processes and establishes a database. Specifically, the spectral confocal sensor can be used to calibrate (extract) the refractive indices of light of multiple different wavelengths propagating in materials of different types (substrates), different thicknesses, and different concentrations (liquid), and fit the dispersion equations in various situations.

[0113] The calibration is divided into single-layer medium calibration and double-layer medium calibration. A mirror is set below the measured medium as a calibration reference. The single-layer medium calibration calculates the refractive index of the single-layer medium (substrate) based on the change in the measurement results of light with a specific wavelength with and without the medium (substrate). By fitting the refractive index curves of the medium with different wavelengths of light and the measured material properties, the dispersion curve of this medium is obtained: n = f(λ, h, ρ, T), where n is the refractive index of the material under different conditions, f is the fitted dispersion equation, λ is the wavelength of the light incident on the medium, h is the thickness of the incident medium, ρ is the density of the medium (including density for solids and considering concentration and density for liquids), and T is the ambient temperature. Thus, the refractive indices of the substrate media corresponding to different wavelengths of light are used for curve fitting, and the corresponding equation is used as the dispersion equation of this substrate medium, denoted as n b (λ).

[0114] The double-layer medium calibration method is used to measure and calibrate materials that cannot exist independently and need to be carried by a substrate. First, the single-layer medium calibration of the substrate is carried out to obtain the dispersion equation and the substrate thickness (known quantity) of the substrate medium. Regarding the double-layer measurement medium as a whole structure, which is called the double-layer measurement medium structure, and using the same method as the single-layer medium calibration, measure x 1 ′, x 2 ′, x 3 ′, and then calculate the refractive index of the thin film layer: Curve fitting is carried out using the refractive indices of the thin films corresponding to different wavelengths of light, and the corresponding obtained equation is used as the dispersion equation (n f (λ)) of the thin film. The refractive indices of different types of thin films (thin film materials) are modeled to provide an optical data source for subsequent film layer thickness measurement, and can compensate for the refractive index during the online measurement process, so as to obtain a higher-precision measurement result.

[0115] For a line spectrum interference sensor, its measurement range is suitable for film layer thickness measurement. For this scenario, this case proposes a method for correcting the thickness of large-area thin films of a spectrum interference sensor. As a preferred implementation, when using a spectrum interference sensor to detect the film thickness at multiple measurement points of a thin film to be measured located on a substrate, a spectrum confocal sensor is mounted on the moving axis of the spectrum interference sensor to synchronously measure the height information at multiple points on the substrate, and the measured film thickness samples are corrected as follows:

[0116] Based on the height information at multiple points on the substrate and combined with the air bearing system parameters, reconstruct the height of the upper surface of the substrate;

[0117] According to the reconstructed height information of the upper surface of the substrate and the dispersion equation of the thin film to be measured, correct the film thickness at each measurement point of the thin film to be measured. The correction method is:

[0118] t f (x, y) = s f (x, y) / (n f (λ(h f (x, y))))

[0119] In the formula, s f (x, y) represents the measured value of the sensor head for the film thickness at the film upper surface coordinates (x, y), h f (x, y) represents the height at the film upper surface coordinates (x, y), λ(h f (x, y)) represents the light wavelength irradiated at the film upper surface coordinates (x, y); n f (λ(h f (x, y))) represents the refractive index at the film upper surface coordinates (x, y).

[0120] The spectral sensor includes three types of sensors, namely, a line spectral interference sensor, a line spectral confocal sensor, and a point spectral confocal sensor. The line spectral interference sensor can achieve on-line thickness value measurement, so that the thickness of the film can be obtained during the scanning of a large-area multi-layer transparent film, and finally the thickness distribution of the large-area film can be used as the main tool for film thickness measurement and film uniformity evaluation; the line spectral confocal sensor can achieve on-line measurement of the film edge morphology and more refined measurement of the thickness distribution measurement area of the line spectral interference sensor, and can quickly obtain the film edge morphology and more accurately locate and construct the defects, and can be used as the main tool for collecting key detail information of the film; the point spectral confocal sensor realizes the detection of the substrate height, and at the same time completes the reconstruction of the substrate surface morphology, and assists the line spectral interference sensor to perform error compensation to achieve high-precision measurement of the film surface morphology in a large area.

[0121] As a specific implementation, for example Figure 4 As shown, by mounting a smaller number of sensors and adding motion axes, the flexibility of the scanning system can be enhanced. The entire surface of the film can be measured in at least five steps with the least number of sensors. First, input the printed substrate, and complete the thickness distribution measurement of half of the film area and the edge morphology measurement in the unilateral substrate movement direction; move the axis to the line spectral confocal sensor to complete the edge morphology scanning in the unilateral main axis direction; move the substrate to achieve the edge scanning of the other axis, and at the same time move the spectral interference sensor and the edge scanning data of the substrate movement direction spectral confocal sensor to the second scanning position; move the substrate back to complete the whole surface spectral interference measurement and the edge scanning in the substrate movement direction; combine the edge scanning data in the substrate movement direction, the edge scanning data in the main axis movement direction, and the spectral interference center scanning data to complete the reconstruction of the whole surface film morphology. According to this set of scanning steps, this set of detection systems can be set after the UV curing chamber, or an additional chamber can be set between the printing chamber and the curing chamber to realize the measurement of dry and wet films through the measurement before and after curing respectively.

[0122] For the line spectral confocal sensor, in principle, since it is difficult for the sensor to receive the interface reflection signal with the refractive index of the film and the substrate being close, in this case, a method for correcting the film thickness measurement of the spectral confocal sensor is proposed. That is, as a preferred implementation, when using the spectral confocal sensor to detect the film thickness at multiple measurement points of the current film to be measured on the substrate, the method for correcting the measured film thickness sample is as follows:

[0123] Based on the current substrate thickness, refractive index, and the refractive index of the current film (the index is the film material and type), correct the thickness of the current substrate to obtain the actual thickness of the current substrate, expressed as:

[0124]

[0125] where t b (x, y) is the actual thickness at the substrate scan line coordinate (x, y), and α b (x, y) is the incident angle of the substrate, s b (x, y) is the measured thickness at the substrate scan line coordinate (x, y), and n b is the refractive index of the substrate, and λ(h b (x, y)) is the light wavelength corresponding to the measured height at the top of the substrate at the substrate coordinate (x, y). By combining the substrate dispersion equation, the refractive index of the incident light in the substrate is obtained, the measurement error caused by height fluctuation is corrected, and the deviation of the measured value caused by the refractive index is corrected. Finally, the true value of the substrate thickness measurement at the position (x, y) is obtained;

[0126] Based on the actual thickness of the substrate, the film thickness at the film measurement point is corrected:

[0127]

[0128] where t f (x, y) is the actual thickness at the film coordinate (x, y), and α f (x, y) is the incident angle of the light beam on the film, s f (x, y) is the measured thickness at the film coordinate x, and λ(g f (x, y)) is the light wavelength corresponding to the measured height at the upper surface of the film at the film coordinate (x, y); n f is the refractive index of the film, and n b is the refractive index of the substrate, and n f and n b are determined based on λ(h f (x, y)) and the dispersion equation; t b (x, y) is the actual thickness at the current substrate coordinate (x, y), and its value is obtained by correcting the current substrate thickness based on the current substrate thickness and refractive index and the refractive index of the current film.

[0129] Specifically, as Figure 5 shown, the sensor obtains the measured value s film (x) of the film plus substrate thickness, the measured value h film (x) of the film surface height, and the measured value s base (x) of the substrate thickness. Due to the refraction of light in the medium, there is a deviation between the measured thickness and distance and the actual value. By combining the fitting corresponding equation of the Z-direction position and the spectral wavelength under the confocal principle and the dispersion equation of the refractive index and wavelength calibrated in the previous process, the actual value of the thickness measurement can be initially obtained through the formula described above.

[0130] As a preferred embodiment, based on the film thickness of each measurement point of the corrected current liquid film, the thickness uniformity of the liquid film in a large area is quantified, and based on the film thickness of each measurement point of the corrected current cured film, the thickness uniformity of the cured film in a large area is quantified. The method for quantifying the film thickness uniformity of the currently measured thin film is as follows:

[0131]

[0132] In the formula, U is the quantified value of film thickness uniformity. If it is greater than the target threshold, the film thickness meets the uniformity index requirement, and t f is the new film thickness sample set of the currently measured thin film.

[0133] As a preferred embodiment, the method further includes: according to the climbing height or climbing distance at the edge of each effective measurement area at the edge, measuring whether the edge topography meets the production requirements. The calculation formula of the measurement standard is: Among them, ΔH is the evaluation index of edge topography height, Δh is the climbing height of the film edge, and h f is the average value of the film thickness samples in the new liquid film thickness sample set or the new cured film thickness sample set.

[0134] Generally speaking, in an inkjet printing system, inkjet printing is carried out on an air-floating conveying device. Before the printing process, optical parameters and thickness calibration of the substrate are collected at multiple points through a spectral reflection sensor. During the printing process, the area of the printed thin film is obtained through the input printing setting information, the thickness information of the printed liquid film is pre-evaluated, and a filtering and compensation method for measurement outliers of the liquid film and the cured film is established. After printing is completed, a spectral interference sensor is used to obtain the thickness distribution of the liquid film before leveling and curing and the thickness distribution of the cured film after leveling and curing respectively to judge whether the composition of each point of the thin film is uniform; a spectral confocal sensor is used to scan each edge of the thin film to obtain the edge topography and thickness characteristics of the thin film and judge whether each thin film edge characteristic parameter meets the index. The present invention is a non-contact and non-destructive on-line thin film quality detection scheme, which combines multi-spectral sensors for thin film measurement and can meet the requirements of the development of inkjet printing towards larger areas and higher efficiency industrialization.

[0135] Embodiment 2

[0136] A large-size substrate inkjet printing thin film encapsulation on-line quality detection system for implementing the large-size substrate inkjet printing thin film encapsulation on-line quality detection method described in Embodiment 1 above, including: an on-line quality detection module and a main control processor;

[0137] The on-line quality detection module includes a thin film internal thickness distribution measurement sub-module and a printing edge topography detection sub-module;

[0138] The main control processor is used to control the internal film thickness distribution measurement sub-module to measure the film thickness through a spectral sensor, control the printed edge topography detection sub-module to measure the edge topography through a spectral confocal sensor, and perform film thickness correction, film thickness uniformity characterization, edge topography correction, and edge topography evaluation processing based on the measurement data.

[0139] As a specific implementation, as Figure 6 shown, the detection system may include: an in-line quality detection chamber and a main control processor; a substrate air-floating conveying module, an internal film thickness distribution measurement module, and a printed edge topography detection module are arranged in the in-line quality detection chamber;

[0140] The main control processor is used to control the air-floating conveying module to transfer the substrate between each cavity, control the air-floating height of the substrate to ensure that the range requirements for subsequent sensor measurement are met, cooperate with the above-mentioned multi-spectral sensor in-line quality detection module, perform the above-mentioned detection of the film thickness uniformity and edge topography of the printed film, multi-sensor data fusion and error compensation, determine whether the film thickness uniformity, edge ramp height, edge ramp distance, edge linearity, etc. of the produced film meet the requirements, and guide the parameter regulation of the production process and the optimization of film performance.

[0141] The multi-spectral sensor group in-line detection module uses a line spectrum interference sensor to measure the film thickness uniformity in the center of the film; the spectral sensor group in-line detection module uses a line spectrum confocal sensor to measure the edge topography of the film.

[0142] As a preferred solution, the substrate air-floating conveying module includes: a first motion component 11 with X1 degree of freedom, a first motion control component 12, an air-floating platform 13, and a first pressure control component 14; the displacement component carries the substrate to move in the X direction on the air-floating platform; the first motion control component is used to control the positive and negative pressure switch actions of the displacement component and the air-floating platform; the air-floating platform 13 is used to suspend the substrate at a certain height when air-floating and conveying the substrate; the first pressure control component 14 is used to control the magnitude of the positive and negative pressure generated by the air-floating platform during air-floating and conveying, so as to control the topography during the substrate transmission process and ensure the smoothness of the substrate transmission.

[0143] The film internal film thickness detection module includes: a point spectrum confocal measurement head 21, a first control and analysis component 22 of the point spectrum confocal sensor, a second motion component 23 with Y1 degree of freedom, a line spectrum interference measurement head 24 fixed to the second motion component, a second control and analysis component 25 of the line spectrum interference sensor, and a second motion control component 26: The measurement end of the point spectrum confocal sensor is fixed in a fixed space near the second motion component and is arranged downward in space; the measurement end of the point spectrum confocal sensor detects the height on both sides of the substrate 42 to be measured in space; the first control and analysis component 22 is connected to the point spectrum confocal measurement head 21 through a signal line, controls the film height measurement according to the trigger signal, collects, analyzes and processes the signal and inputs it to the main controller 5. The second control and analysis component 25 is connected to the line spectrum interference measurement head 24 through a signal line. The main controller 5 sends a signal to the second control and analysis component, controls the line spectrum interference measurement head 24 to complete the measurement of the thickness distribution of the film 41 to be measured according to the signal, collects, processes and analyzes the signal and inputs it to the main controller 5; the second motion component 23 completes the movement of the line spectrum interference sensor, performs the Y switching movement of the measurement point after the substrate 42 to be measured is removed, and after the substrate returns to the cavity, completes the measurement of the film thickness distribution of the entire surface. The second motion control component 26 is used to control the switching movement of the measurement position of the sensor head after the point spectrum confocal receives the substrate departure signal, and completes the measurement process.

[0144] The film edge morphology detection module includes: a third motion component 31 with Y2 degree of freedom, a fourth motion component 33 with Y3 degree of freedom, a line spectral confocal measurement head 32 fixed on the third motion component 31 and the fourth motion component 33, a third motion control component 34 of the third motion component 31, a third control and analysis component 35 of the line spectral confocal sensor, and a fourth motion control component 36 of the fourth motion component 33; the line spectral confocal measurement head 32 is arranged downward in space, wherein the line spectral confocal measurement head 32 installed on the third control component 31 is used to measure the film edge on the Y direction side, and the line spectral confocal measurement head 32 installed on the fourth control component 33 is used to measure the film edge on the X direction side; the third motion component 31 and the fourth motion component 33 are used as mechanical components to realize the movement of the sensor probe 32 in the Y direction. Among them, the third motion component 31 is used to switch the measurement position on one side of the Y direction of the edge, and the fourth motion component 33 is used to scan the film edge on the X direction side. The third motion component 31 and the fourth motion component 33 are respectively controlled by the third motion control component 35 and the fourth motion control component 36. The control signal for the movement process is received by the main control processor 5 from the signal returned by the point spectral confocal measurement head 21 to the second control and analysis component 22 as an indication, and the main control processor 5 sends a motion signal to the third and fourth motion control components 35 and 36 to execute the corresponding motion; the third control and analysis component 35 is connected to the line spectral confocal measurement head 32 through a signal line, controls the line spectral confocal measurement head to collect the film edge morphology according to the starting signal, analyzes and processes the collected signal, and inputs it to the main control processor 5.

[0145] The substrate air-floating conveying module cooperates with the film multi-spectral sensor film thickness and edge detection module to convey the substrate to be measured of the printed film to the specified position, and completes the detection operation of the substrate by means of line scanning of the moving substrate and scanning of the moving sensor.

[0146] The related technical solutions are the same as those in Embodiment 1 and will not be elaborated here.

[0147] Those skilled in the art can easily understand that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for online quality inspection of inkjet printed thin film packaging on large-size substrates, characterized in that: include: The spectral sensor is used to measure the thickness of multiple measuring points of the current film to be measured online to form a film thickness sample set. The Z-score of the film thickness of each measuring point is calculated. When the Z-score is greater than the threshold, the film thickness average of the two measuring points adjacent to the measuring point on the film surface space is used to replace the film thickness data of the measuring point in the film thickness sample set to form a new film thickness sample set. Obtain the dispersion equation corresponding to the current film to be measured from the pre-built database, and determine the wavelength of light irradiated by the sensor to each measuring point on the upper surface of the film; obtain the film refractive index at the corresponding measuring point from the wavelength of light and the dispersion equation; calculate the optical path difference of the sensor light beam in the film based on the film thickness sample value of each measuring point in the new film thickness sample set, and based on the relationship between the optical path difference and the film refractive index, combined with the film refractive index at the corresponding measuring point, calibrate and obtain the true value of the film at the corresponding measuring point to quantify the film thickness uniformity of the current film to be measured; The spectral confocal sensor is used to measure the height of multiple measuring points on the edge of the film to be measured online and calibrate them to obtain the true morphology of the edge; wherein the multiple measuring points of the edge are located in the four preset effective edge measurement areas, and the coordinates of the upper left corner are (a1-n, b1+n), (a2-s, b1+n), (a1+s, b2+s), (a1+s, b1+n), and the coordinates of the lower right corner are (a1+s, b2-n), (a2+n, b2-n), (a2-s, b2-n), (a2-s, b1-s); wherein, (a1, b1) and (a2, b2) are the coordinates of the upper left corner and the lower right corner of the printing domain determined based on the printing vector element information, s = α*(m%d), n = ds, s represents the maximum length of the edge of the film in the height measurement field when the film first appears, m represents the estimated maximum climbing distance of the film, d represents the line width of the field of view of a single measurement of the spectral confocal sensor, α is the preset proportional coefficient, and n represents the length occupied by the substrate in the corresponding height measurement field.

2. A method for online quality inspection of inkjet printed thin film packaging of large-size substrates as claimed in claim 1, characterized in that: When a line spectrum interferometer is used to measure the thickness of multiple measuring points in the central effective area of ​​the film to be measured, the measuring point area is the central effective rectangular area determined based on the known printing primitive vector data information, which is expressed as: Where, (a1, b1) and (a2, b2) are the coordinates of the upper left corner and lower right corner of the print area determined based on the print vector primitive information, (a1′, b1′) and (a2′, b2′) are the coordinates of the upper left corner and lower right corner of the central effective rectangular area, respectively. a The square root of the total number of preset measuring points, n v The value is the square root of the known number of valid measuring points.

3. A method for online quality inspection of inkjet printed thin film packaging of large-size substrates as claimed in claim 2, characterized in that: When the spectral confocal sensor is used to measure the thickness of multiple measuring points in the four edge effective measurement areas of the current film to be measured, it also includes: fusing the film thickness of the measuring points that overlap the central effective rectangular area and the four edge effective measurement areas, and using the integrated thickness measurement value after the fusion processing as the film thickness measurement value of the corresponding measuring point to perform Z-score calculation, wherein the fusion processing method is: Among them, t ic is the comprehensive thickness measurement value, t c is the thickness measurement value of the spectral confocal sensor at the coincident measurement point, t i is the thickness measurement value of the spectral sensor at the coincident measuring point, σ c is the standard deviation of the thickness measurement value of the spectral confocal sensor, σ i is the standard deviation of the thickness measurement value of the spectral sensor.

4. A method for online quality inspection of inkjet printed thin film packaging of large-size substrates as claimed in claim 1, characterized in that: The film to be tested includes a liquid film and a cured film, and corresponds to a new liquid film thickness sample set and a new cured film thickness sample set respectively; The estimated value of the printed film thickness is calculated using the printing input parameters as the theoretical mean value of the liquid film thickness for calculating the Z-score of the liquid film thickness sample; the known target printed film thickness value is used as the theoretical mean value of the cured film thickness for calculating the Z-score of the cured film thickness sample; The Z-score of the liquid film thickness at each measuring point is: In the formula, is the Z-score value of the i-th measuring point of the liquid film, is the measured value of the liquid film thickness at the i-th measuring point of the liquid film, is the mean value of the liquid film thickness at all measuring points, and m is the number of liquid film thickness measuring points; is the estimated value of the printing liquid film thickness, n p 、n e are the number of printed pixels determined based on the known printing primitive vector data information and the number of extracted film edge pixels, respectively. is the known average printing droplet volume, S is the printing film area determined based on the edge information of the known printing primitive vector data; The Z-score of the cured film thickness at each measuring point is: In the formula, is the Z-score of the i-th measuring point of the cured film, is the measured value of the cured film thickness at the i-th measuring point of the cured film, is the theoretical mean value of the cured film thickness, is the mean value of the cured film thickness of all measuring points, and m is the number of measuring points for the cured film thickness.

5. The method for online quality inspection of inkjet printed thin film packaging of large-size substrates as claimed in claim 1, characterized in that: The dispersion equations corresponding to various films in the database are constructed as follows: The stacked structure of the target film and the target substrate is regarded as a double-layer measuring medium structure. Under the detection light of different wavelengths λ, the distance x1′ from the sensor head to the upper surface of the double-layer measuring medium structure, the distance x2′ from the sensor head to the lower surface of the double-layer measuring medium structure, and the distance x3′ from the sensor head to the upper surface of the reflector are measured offline. The sensor is set just above the double-layer measuring medium structure, and the reflector is set just below the double-layer measuring medium structure. Based on n b (λ) and the target substrate thickness H, combined with the geometric relationship of beam propagation, calculate the refractive index of the target film for each wavelength of light to fit the dispersion equation n of the target film f (λ), where n b (λ) represents the dispersion equation of the target substrate medium; The refractive index of the target film for each wavelength of light is calculated by the following formula: Where n f，λ is the refractive index of the target film for light with a wavelength of λ, θ2 is the refraction angle of light propagating in the film layer, and θ3 is the refraction angle of light propagating in the substrate.

6. A method for online quality inspection of inkjet printed thin film packaging of large-size substrates as claimed in claim 4, characterized in that: When using a spectral interferometer sensor to detect the film thickness at multiple measuring points of a thin film to be measured on a substrate, a spectral confocal sensor is mounted on the moving axis of the spectral interferometer sensor to synchronously measure the height information at multiple points of the substrate, and the measured film thickness samples are corrected as follows: Based on the height information of multiple points on the substrate and combined with the parameters of the air flotation system, the height of the upper surface of the substrate is reconstructed; According to the reconstructed substrate surface height information and the dispersion equation of the film to be measured, the film thickness of each measuring point of the film to be measured is corrected in the following way: t f (x,y)=s f (x,y) / (n f (λ(h f (x,y)))) In the formula, s f (x, y) represents the measurement value of the film thickness by the sensor head at the coordinate (x, y) on the film surface, h f (x, y) represents the height of the film surface at coordinate (x, y), λ(h f (x, y)) represents the wavelength of light irradiated to the surface coordinate (x, y) of the film; n f (λ(h f (x, y))) represents the refractive index at the coordinate (x, y) on the surface of the film.

7. A method for online quality inspection of inkjet printed thin film packaging of large-size substrates as claimed in claim 4, characterized in that: When a spectral confocal sensor is used to detect the film thickness of multiple measuring points of the current film to be measured on the substrate, the method of correcting the measured film thickness sample is as follows: Based on the current substrate thickness and refractive index and the refractive index of the current film (the index is the film material and type), the thickness of the current substrate is corrected to obtain the actual thickness of the current substrate; Based on the actual thickness of the substrate, the film thickness of the thin film measuring point is corrected: Among them, t f (x, y) is the actual thickness of the film at coordinate , α f (x, y) is the angle of the measuring beam incident on the film, s f (x, y) is the measured thickness of the film at coordinate x, λ(h f (x, y)) is the wavelength of light corresponding to the measured height at the coordinate (x, y) on the film surface; n f is the refractive index of the film, n b is the refractive index of the substrate, n f and n b The value of is based on λ(h f (x, y)) and the dispersion equation; t b (x, y) is the actual thickness at the current substrate coordinate (x, y), and its value is obtained by correcting the current substrate thickness based on the current substrate thickness and refractive index and the refractive index of the current film.

8. The method for online quality inspection of inkjet printed thin film packaging of large-size substrates as claimed in claim 1, characterized in that: The method to quantify the uniformity of the film thickness of the film to be tested is: Where U is the quantitative value of film thickness uniformity. If it is greater than the target threshold, the film thickness meets the uniformity index requirements. f It is a new film thickness sample set of the film to be tested.

9. The method for online quality inspection of inkjet printed thin film packaging of large-size substrates as claimed in claim 1, characterized in that: Also includes: According to the climbing height or climbing distance of the edge of each effective measurement area, measure whether the edge morphology meets the production requirements. The calculation formula of the measurement standard is: Among them, ΔH is the edge morphology height evaluation index, Δh is the climbing height of the film edge, and h f It is the mean value of the film thickness samples in the new liquid film thickness sample set or the new cured film thickness sample set.

10. An online quality inspection system for large-size substrate inkjet printing thin film encapsulation, characterized in that: Used to perform an online quality detection method for inkjet printing thin film packaging of a large-size substrate as claimed in any one of claims 1 to 9, comprising: an online quality detection module and a main control processor; The online quality detection module includes a film internal thickness distribution measurement submodule and a printing edge morphology detection submodule; The main control processor is used to control the film internal thickness distribution measurement submodule to measure the film thickness through a spectral sensor, control the printing edge morphology detection submodule to measure the edge morphology through a spectral confocal sensor, and perform film thickness correction, film thickness uniformity characterization, edge morphology correction and edge morphology evaluation processing based on the measurement data.

Images