A metal film size measurement method and system based on optical measurement

Abstract

The present invention relates to the field of optical measurement technology, and discloses a metal film size measurement method and system based on optical measurement, comprising: irradiating a metal film with a multi-wavelength laser to generate an interference signal; modulating the wavelength of the incident light, and adjusting the attention of different wavelengths respectively through a sliding time window; obtaining interference fringes in each time window, and obtaining the measurement result of each time window based on the interference fringes; performing cluster selection based on the measurement result of the time window, and re-measuring using the selection result to obtain the final measurement result. The method can effectively improve the measurement accuracy and stability of the metal film size. By dynamically adjusting the attention of different wavelengths and combining the cluster selection and re-measurement process to select a wavelength sensitive to the parameters, the reliability of the final measurement result is guaranteed, and the high requirements of modern precision manufacturing and quality control for film size measurement are met.

Description

Technical Field

[0001] The present invention relates to the field of optical measurement technology, and in particular to a metal film size measurement method and system based on optical measurement. Background Art

[0002] With the widespread application of metal thin films in electronics, optics, and semiconductors, the demand for precise dimensional measurement continues to increase. Traditional dimensional measurement methods, such as contact probes and scanning electron microscopy (SEM), while highly accurate, often suffer from issues such as contact damage, slow measurement speeds, and high equipment costs. Optical measurement technology, particularly non-contact measurement methods based on interferometry, has become an increasingly important tool for thin film dimensional measurement due to its high precision, speed, and non-destructive nature.

[0003] Currently, there are existing methods for measuring film thickness based on interferometry. However, due to the varying sensitivity of interference fringes at different wavelengths to film dimensions, and the potential for signal noise and multiple interferences in practical applications, the measurement accuracy of a single wavelength is limited. Therefore, utilizing multi-wavelength laser light sources, along with appropriate interference fringe analysis and data processing techniques, to improve the combined measurement accuracy of the length, width, and thickness of metal films has become an important area of ​​technological development. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problems solved by the present invention are: the measurement accuracy of a single wavelength is limited, the interference fringes of different wavelengths have different sensitivities to the film size, and in practical applications, problems such as signal noise and multiple interference may exist.

[0006] To solve the above technical problems, the present invention provides the following technical solution: a method for measuring the size of a metal film based on optical measurement, comprising:

[0007] The metal film is irradiated by a multi-wavelength laser to generate an interference signal;

[0008] Modulate the wavelength of the incident light and adjust the attention to different wavelengths by sliding the time window;

[0009] Obtaining interference fringes in each time window, and obtaining a measurement result for each time window based on the interference fringes;

[0010] Cluster selection is performed according to the measurement results of the time window, and re-measurement is performed using the selection results to obtain the final measurement results.

[0011] As a preferred embodiment of the metal film size measurement method based on optical measurement described in the present invention, the multi-wavelength laser comprises randomly selecting N laser sources of different wavelengths to form an incident light combination;

[0012] The selected multi-wavelength laser source is irradiated onto the metal film. The laser beam interacts with the surface. A high-resolution CCD or CMOS camera is used to receive the reflected light and capture the interference images at different wavelengths to obtain the interference fringe pattern generated by each wavelength of laser.

[0013] As a preferred embodiment of the optical measurement-based metal film dimension measurement method of the present invention, the modulation of the wavelength of the incident light comprises weighting the analysis results of N different wavelengths;

[0014] Randomly select a wavelength from N interference signals with different wavelengths and set it as m;

[0015] After obtaining the interference fringe pattern generated by the laser of each wavelength, the interference fringe pattern corresponding to each wavelength is analyzed, and the analysis results of each wavelength are integrated;

[0016] During integration, the parsing results corresponding to wavelength m are subjected to attention weighting processing;

[0017] The analysis process includes obtaining the surface morphological characteristics of the film by performing surface fitting on the edge position and morphology of the interference fringes; and obtaining thickness information by calculating the spacing of the interference fringes, the thickness of the film, the refractive index, the angle of incidence, and the wavelength of the light wave.

[0018] As a preferred embodiment of the optical measurement-based metal film dimension measurement method of the present invention, the adjustment of attention to different wavelengths by sliding the time window includes setting a fixed time window length, randomly selecting a wavelength in each time window, and performing attention weighting processing on the corresponding analysis results;

[0019] The constraint when selecting the wavelength is that the selected wavelengths in N consecutive time windows are all different.

[0020] As a preferred solution of the metal film size measurement method based on optical measurement according to the present invention, the measurement result of each time window includes weighting the analysis results corresponding to N wavelengths in each time window; assuming that: Represents the weight coefficient. If n corresponds to the analytical result corresponding to the unweighted wavelength, then ; If n corresponds to the analytical result corresponding to the weighted wavelength, then ; M represents the weighting coefficient, and M>2; i∈{C, K, G}, C represents the length of the metal film, K represents the width of the metal film, and G represents the thickness of the metal film;

[0021] For each parameter i of the metal film, M analytical results corresponding to the weighted wavelength are copied according to the weight coefficient, and a parameter set is formed with the analytical results corresponding to other N-1 wavelengths as the measurement result.

[0022] As a preferred embodiment of the optical measurement-based metal film dimension measurement method of the present invention, the cluster selection based on the measurement results of the time window includes obtaining N consecutive time windows from the current time series, and performing cluster selection for each parameter i of the metal film in any time window;

[0023] The cluster selection includes clustering the measurement results. If there is only one cluster center, the wavelength of the current window attention weight is determined. For the measurement results of the metal film parameter i, The criterion is satisfied; otherwise, it is judged that the criterion is not satisfied;

[0024] The re-measurement includes removing k wavelengths that do not meet the criteria from N different wavelengths, and if the number of remaining wavelengths is less than N / 2, randomly selecting k other wavelengths, performing interference fringe analysis and cluster selection on the re-selected wavelengths in a sliding time window until the number of remaining wavelengths after removing the wavelengths that do not meet the criteria is not less than N / 2; obtaining and outputting analysis results for the wavelengths that meet the criteria with respect to the parameter i;

[0025] The k other wavelengths do not include the removed wavelengths; the reselected wavelengths include the remaining wavelengths that meet the criteria and the newly added k other wavelengths.

[0026] As a preferred embodiment of the optical measurement-based metal film dimension measurement method of the present invention, the final measurement result includes cluster selection and re-measurement of the two parameters C and K at the same time; after the output is completed, cluster selection and re-measurement of the parameter G are performed;

[0027] When all three parameters have completed cluster selection and re-measurement and output, the test process ends;

[0028] averaging the analytical results of the parameter i corresponding to the wavelengths that meet the criterion to obtain the final measurement result;

[0029] The process of randomly selecting multiple wavelengths includes the following constraints:

[0030] N>3;

[0031] The maximum difference of multiple randomly selected wavelengths is not less than E;

[0032] Wherein, E represents the threshold value of the maximum wavelength difference.

[0033] A metal film size measurement system based on optical measurement using any method described in the present invention, characterized in that:

[0034] The acquisition unit irradiates the metal film with a multi-wavelength laser to generate an interference signal;

[0035] The attention control unit modulates the wavelength of the incident light and adjusts the attention of different wavelengths by sliding the time window;

[0036] a measuring unit, which obtains interference fringes in each time window and obtains a measurement result of each time window based on the interference fringes;

[0037] The output unit performs cluster selection according to the measurement results of the time window, and remeasures using the selection results until the criterion is met to obtain the final measurement result.

[0038] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein: when the processor executes the computer program, the steps of any one of the methods of the present invention are implemented.

[0039] A computer-readable storage medium stores a computer program, wherein: when the computer program is executed by a processor, the steps of any one of the methods of the present invention are implemented.

[0040] The beneficial effects of this invention are as follows: The optical measurement method for metal film dimensions provided by this invention utilizes multi-wavelength laser interferometry technology, combined with a sliding time window and weighted analysis methods, to effectively improve the measurement accuracy and stability of metal film dimensions (length, width, and thickness). By dynamically adjusting the focus of different wavelengths and combining cluster selection and re-measurement to select wavelengths sensitive to parameters, the reliability of the final measurement results is ensured, providing high adaptability and accuracy, meeting the stringent requirements for thin film dimension measurement in modern precision manufacturing and quality control. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1This is an overall flow chart of a metal film size measurement method based on optical measurement provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0043] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0044] Example 1, reference Figure 1 , as one embodiment of the present invention, provides a metal film size measurement method based on optical measurement, comprising:

[0045] S1: Irradiate the metal film with a multi-wavelength laser to generate an interference signal.

[0046] The multi-wavelength laser system includes randomly selecting N laser sources of different wavelengths to form a combination of incident light. The selected multi-wavelength laser sources are then irradiated onto the metal film. The laser beams interact with the surface, and a high-resolution CCD or CMOS camera receives the reflected light. Interference images at different wavelengths are captured, yielding the interference fringe pattern generated by each wavelength of laser light.

[0047] When a laser beam strikes the surface of a metal film, the reflected light interferes with the light reflected back from the film. At the film's surface, the phase and path difference of the light waves vary due to the film's thickness and surface shape, leading to constructive and destructive interference between the light waves. Different wavelengths of light produce different spacing of interference fringes. Longer wavelengths produce wider-spaced interference fringes, while shorter wavelengths produce denser fringes. This enables the generation of multiple interference patterns using a multi-wavelength laser source. The interference fringes generated by each wavelength are spatially superimposed to form a composite interference pattern. After being scattered or reflected by the film's surface, the reflected light enters a high-resolution CCD or CMOS camera. These cameras capture interference images at different wavelengths, recording the intensity distribution and phase variations of the reflected light. The interference pattern corresponding to each wavelength appears as periodic fringes or alternating patterns of light and dark. The camera transmits the captured images to a computer for further processing. Image processing algorithms (such as edge detection and Fourier analysis) can be used to extract information such as the position, spacing, and shape of the interference fringes from the captured interference pattern.

[0048] The interference pattern for each wavelength exhibits a unique fringe pattern, which can be straight, curved, or circular, depending on the laser's incident angle, the film's surface morphology, and its refractive index. The spacing of the interference fringes is closely related to the film's thickness, refractive index, incident angle, and laser wavelength. By measuring the spacing of these fringes, the film's thickness and other geometric characteristics can be inferred.

[0049] Composite interference patterns contain multiple frequency components. Extracting a single-wavelength pattern from a composite interference pattern primarily relies on image processing techniques such as Fourier transforms, frequency-domain filtering, and fringe period analysis. By analyzing the frequency components, period, and phase differences of the interference fringes, the interference fringes of each wavelength can be effectively separated from the composite pattern, thereby extracting the single-wavelength interference pattern. Fourier transforms are an effective frequency-domain analysis method that converts spatial information in an image into frequency-domain information, helping to separate interference fringes generated by different wavelengths.

[0050] S2: Modulate the wavelength of the incident light and adjust the attention of different wavelengths separately through a sliding time window.

[0051] The modulation of the wavelength of the incident light includes weighting the analysis results of N different wavelengths. Randomly select a wavelength from the interference signals of N different wavelengths and set it as m. After obtaining the interference fringe pattern generated by the laser of each wavelength, the interference fringe pattern corresponding to each wavelength is analyzed, and the analysis results of each wavelength are integrated. During the integration, the analysis results corresponding to the wavelength m are weighted. The analysis process includes obtaining the surface morphological characteristics of the film by performing surface fitting on the edge position and morphology of the interference fringes; and obtaining thickness information by calculating using the spacing of the interference fringes, the thickness of the film, the refractive index, the angle of incidence, and the wavelength of the light wave.

[0052] It is important to note that by combining interference signals from multiple wavelengths, the limitations of single-wavelength lasers can be overcome and measurement errors can be reduced. Interference fringes from different wavelengths have different sensitivities under different conditions, which can provide more measurement information.

[0053] It's important to note that in multi-wavelength laser interferometry, each wavelength responds differently to the film's reflection and interference signals. Therefore, the interference fringes corresponding to different wavelengths may have varying sensitivities to different dimensional features of the film. By randomly selecting and weighting different wavelengths for analysis, the advantages of each wavelength can be fully utilized, avoiding measurement errors associated with a single wavelength for certain dimensions or surface features, thereby improving the overall accuracy of the measurement results.

[0054] Interference patterns at different wavelengths may be affected by different degrees of noise. By weighting the analytical results of different wavelengths, especially giving more attention to certain wavelengths, the contribution of the more reliable measurement results of those wavelengths is enhanced, thus making it a cluster center.

[0055] Adjusting attention to different wavelengths using a sliding time window involves setting a fixed time window length, randomly selecting a wavelength in each time window, and performing attention weighting on the corresponding analysis results. The constraint on wavelength selection is that the selected wavelengths in N consecutive time windows are all different.

[0056] It should be noted that interference signals at different wavelengths may have different sensitivities to parameters such as film thickness and surface morphology. By sliding the time window and selecting different wavelengths, the effect of each time window is enhanced, and its reliability can be inferred based on its prominence.

[0057] S3: Obtain interference fringes in each time window, and obtain a measurement result of each time window based on the interference fringes.

[0058] The measurement result of each time window includes weighting the analysis results corresponding to N wavelengths in each time window; assuming that, Represents the weight coefficient. If n corresponds to the analytical result corresponding to the unweighted wavelength, then ; If n corresponds to the analytical result corresponding to the weighted wavelength, then ; M represents the weighting coefficient, and M>2; i∈{C, K, G}, C represents the length of the metal film, K represents the width of the metal film, and G represents the thickness of the metal film.

[0059] For each parameter i of the metal film, M analytical results corresponding to the weighted wavelength are copied according to the weight coefficient, and a parameter set is formed with the analytical results corresponding to other N-1 wavelengths as the measurement result.

[0060] The point is that the multi-head attention strategy strengthens the features of different wavelengths in different time windows, and by weighting and replicating the analysis results of weighted wavelengths, the influence of these wavelengths in the clustering process is increased. In multi-wavelength optical measurements, different wavelengths have different sensitivities to different film parameters (such as length, width, and thickness). By weighting and replicating the analysis results of different wavelengths within each time window, the influence of different wavelengths is intentionally enhanced, allowing the analysis results of the wavelengths that receive attention to become cluster centers. This facilitates the selection of wavelengths that are sensitive to each parameter during cluster selection.

[0061] S4: performing cluster selection according to the measurement results of the time window, and re-measuring using the selection results to obtain a final measurement result.

[0062] Furthermore, cluster selection based on the measurement results of the time window includes obtaining N consecutive time windows from the current time series, and performing cluster selection for each parameter i of the metal film in any time window. The cluster selection includes clustering the measurement results, and if there is only one cluster center, determining the wavelength of the attention weight of the current window, and the measurement results of the metal film parameter i. The criterion is satisfied; otherwise, it is judged that the criterion is not satisfied (the purpose of this step is to screen wavelengths that are sensitive to parameter i).

[0063] It's important to note that within each time window, the analysis results for different wavelengths are weighted. In particular, attention weighting for a particular wavelength means that the data for that wavelength is considered more important within the current time window. Therefore, the weighted wavelength is more likely to become the "dominant" data in the cluster analysis, while other wavelengths may only participate in the clustering as supplementary data. If the analysis results for a weighted wavelength (a wavelength enhanced by the attention mechanism) are consistent with the results for other wavelengths (i.e., forming a single cluster center), then the data for these weighted wavelengths is consistent with the measurement results for most other wavelengths and can be considered normal and reliable.

[0064] In multi-wavelength optical measurements, each wavelength has a different sensitivity to thin film parameters. The interference fringes generated by different wavelengths during each measurement reflect surface morphological features to varying degrees. Consequently, the analysis results for multiple wavelengths may exhibit varying fluctuations or errors. If the analysis results for a particular wavelength are inconsistent with those for other wavelengths, this may indicate inaccurate or unreliable measurements at that wavelength. Cluster analysis is used to group the measurement results for these wavelengths and determine their reliability by determining whether there is a single cluster center.

[0065] There is only one cluster center: If after cluster analysis, the analysis results of all wavelengths form a cluster center, then these analysis results are consistent with each other and conform to the actual measurement situation, indicating that the analysis results of these wavelengths are relatively accurate in the current measurement.

[0066] Two cluster centers: If there are two cluster centers, it means that the weighted wavelength analysis results deviate from the mainstream results. This may be because the interference signal of some wavelengths is affected by noise, or the wavelength is not sensitive to the parameter, resulting in inconsistent measurement results with other wavelengths. In this case, it is necessary to eliminate the inappropriate wavelengths to avoid affecting the final measurement results.

[0067] This method allows us to select wavelengths that are sensitive only to parameter i. If the data for a particular wavelength is abnormal, or if that wavelength is less sensitive to a particular film parameter, these unreliable wavelengths can be effectively eliminated by removing the wavelengths that do not meet the criteria. This prevents erroneous data from these wavelengths from affecting the final measurement results, ensuring accuracy.

[0068] The re-measurement includes removing k wavelengths that do not meet the criteria from N different wavelengths. If the number of remaining wavelengths is not less than N / 2, the analytical results corresponding to the wavelengths that meet the criteria with respect to parameter i are directly output and output; if the number of remaining wavelengths is less than N / 2, k other wavelengths are randomly selected and interference fringes are analyzed and clustered for the re-selected wavelengths in a sliding time window until the number of remaining wavelengths is not less than N / 2 after the wavelengths that do not meet the criteria are removed (at least N new time windows must pass after the re-measurement begins); the analytical results corresponding to the parameter i for the wavelengths that meet the criteria are obtained and output (the purpose of this step is to ensure that at least half of the wavelengths are sensitive to parameter i, and only the analytical results of the sensitive wavelengths are used for output). The k other wavelengths do not include the wavelengths that have been removed; the re-selected wavelengths include the remaining wavelengths that meet the criteria and the k newly added other wavelengths.

[0069] In multi-wavelength interferometry, in order to ensure the reliability and accuracy of the results, it is necessary to rely on a sufficient number of wavelengths. If, during the preliminary analysis process, after removing wavelengths that do not meet the criteria, the number of remaining wavelengths is less than N / 2, that is, the number of remaining wavelengths is less than half, then these remaining wavelengths may not provide enough valid information for accurate measurement. Therefore, by randomly selecting other wavelengths to supplement until the number of remaining wavelengths is no less than N / 2, it is ensured that there are enough wavelengths involved in the measurement and that these wavelengths can provide sufficient characteristic information, thereby improving the reliability of the final result. This step is to avoid insufficient measurement information due to inaccurate data of certain wavelengths, and to ensure that accurate measurement results can be obtained from sufficient and valid wavelength data.

[0070] Furthermore, the final measurement result includes simultaneous cluster selection and remeasurement of parameters C and K. After the cluster selection and remeasurement are completed and output, parameter G is cluster selected and remeasured. The test process ends when cluster selection and remeasurement are completed and output for all three parameters. The analytical results for parameter i corresponding to the wavelengths that meet the criteria are averaged to obtain the final measurement result.

[0071] During multi-wavelength optical measurement, different wavelengths have varying sensitivities to various parameters. By performing cluster selection and re-measurement in stages, we can focus on optimizing each parameter, ensuring optimal measurement accuracy for each. Specifically, length (C) and width (K) are in-plane dimensions of the film, and their measurements are directly affected by optical interference fringes. These two parameters are typically measured in the same plane, making their measurements independent and unaffected by the thickness parameter (G). By jointly measuring C and K within the same time window and optimizing the results through clustering and re-measurement, we can accurately determine the dimensions of both parameters. Thickness (G), a vertical dimension of the film, is not directly related to the length and width measurements and should therefore be handled separately during the measurement process. By first confirming the accurate values ​​of C and K before beginning thickness measurement, we avoid interference from variations in length and width. This ensures a more accurate thickness measurement that is independent of in-plane dimension variations.

[0072] While length, width, and thickness are both dimensional parameters of thin films, their measurement processes have different optical characteristics. Length and width are planar parameters affected by factors such as the angle of incidence and the distribution of interference fringes, while thickness is more dependent on the perpendicularity of the interference fringes and the interference effect of light. Measuring them separately prevents variations in length and width from affecting thickness measurements, ensuring that the measurement results of each parameter are unaffected by interference from the other.

[0073] The process of randomly selecting multiple wavelengths includes the following constraints:

[0074] N>3; the maximum difference of multiple randomly selected wavelengths is not less than E; where E represents the threshold value of the maximum wavelength difference.

[0075] On the other hand, this embodiment also provides a metal film size measurement system based on optical measurement, which includes:

[0076] The acquisition unit irradiates the metal film with a multi-wavelength laser to generate an interference signal;

[0077] The attention control unit modulates the wavelength of the incident light and adjusts the attention of different wavelengths by sliding the time window;

[0078] a measuring unit, which obtains interference fringes in each time window and obtains a measurement result of each time window based on the interference fringes;

[0079] The output unit performs cluster selection according to the measurement results of the time window, and remeasures using the selection results until the criterion is met to obtain the final measurement result.

[0080] If the above functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0081] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0082] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, and then editing, interpreting, or processing in another suitable manner as necessary, and then storing it in a computer memory.

[0083] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0084] Example 2 is an embodiment of the present invention, which provides a metal film size measurement method based on optical measurement. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0085] The dimensional measurements of metal films were performed using the traditional single-light source method, the invented method, and standard values. The experimental objects were aluminum (Al), copper (Cu), and tin (Sn) films with thicknesses of 100nm, 500nm, and 1000nm, respectively.

[0086] Three metal films of aluminum (Al), copper (Cu) and tin (Sn) were selected, and the film size was 10 mm × 10 mm.

[0087] The thickness of each metal film is 100nm, 500nm and 1000nm respectively.

[0088] Before the experiment, the film surface was cleaned to ensure that there were no surface contaminants.

[0089] Traditional single light sources use lasers with a wavelength of 532nm.

[0090] The present invention uses four lasers with different wavelengths (532nm, 650nm, 780nm and 980nm) for interference measurement and combines a weighted algorithm to optimize the interference signal.

[0091] Standard values ​​are based on measurements made with known standard instruments, used as a reference.

[0092] Interferometric image acquisition:

[0093] The interference fringes of the reflected light are captured using a high-resolution CCD camera.

[0094] Fourier transform and edge detection algorithm are used to extract interference fringe data.

[0095] Data processing:

[0096] The traditional method is to measure the size directly through a single light source.

[0097] The method of the present invention generates interference signals through multi-wavelength lasers and combines them with weighted algorithms for processing to improve measurement accuracy.

[0098] Standard values ​​are based on direct measurement of film dimensions using calibrated precision instruments.

[0099] Data recording and comparison:

[0100] The measurement results under each method were recorded, and the differences between the invented method, the traditional single light source method, and the standard values ​​were compared, as shown in Table 1.

[0101] Table 1 Experimental data table

[0102]

[0103] Based on the data in the table, the advantages of the present invention over the traditional single light source method can be analyzed from the following aspects:

[0104] Measurement accuracy: For the length (length C) and width (width K) measurements of the film, the measurement results obtained using the multi-wavelength weighted method of the present invention (such as aluminum film: 9.87 mm, copper film: 10.03 mm, tin film: 10.07 mm) are very close to the standard values ​​(such as aluminum film: 9.90 mm, copper film: 10.05 mm, tin film: 10.09 mm), and are more accurate than the results of the traditional single light source method (such as aluminum film: 9.85 mm, copper film: 10.01 mm, tin film: 10.05 mm).

[0105] The present invention can effectively reduce the measurement error caused by a single wavelength by weighted processing of signals of different wavelengths. Especially when the film is thick (such as 500nm and 1000nm), the advantage of multi-wavelength processing is more obvious.

[0106] Thickness measurement: In thickness (G) measurement, the results of traditional methods at different wavelengths (such as aluminum film: 1250nm, copper film: 1275nm, tin film: 1300nm) show certain deviations. This may be due to the matching problem between the single wavelength laser and the film properties.

[0107] By weighting interference signals of different wavelengths, the method of the present invention obtains more consistent thickness measurement results (such as aluminum film: 1250nm, copper film: 1278nm, tin film: 1302nm). The measurement results are close to the standard values ​​(such as aluminum film: 1252nm, copper film: 1280nm, tin film: 1304nm), proving the high precision of the method.

[0108] Error Analysis: The traditional single-light source method has large measurement errors, especially when measuring thicker films. For example, the copper film thickness measurement result was 1275nm, while the standard value was 1280nm, with an error of 5nm. However, the result of 1278nm obtained using the method of the present invention was reduced to 2nm from the standard value, demonstrating that the method of the present invention can effectively reduce measurement errors.

[0109] Compared with traditional methods, the present invention can significantly improve the accuracy and consistency of film size measurement through multi-wavelength weighted processing and signal optimization.

[0110] Measurement consistency and reliability: Traditional methods can exhibit some fluctuation in measurement results over multiple experiments, particularly when measuring thicker films. In contrast, the method presented here, through weighted and optimized processing of multi-wavelength data, achieves more consistent measurement results and significantly improves reliability.

[0111] The measurement results of the standard values ​​are very close to those of the method of the present invention, which shows that the present invention can achieve measurement accuracy comparable to that of precision instruments in practical applications.

[0112] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for measuring the size of a metal film based on optical measurement, characterized in that: include: The metal film is irradiated by a multi-wavelength laser to generate an interference signal; Modulate the wavelength of the incident light and adjust the attention to different wavelengths by sliding the time window; Obtaining interference fringes in each time window, and obtaining a measurement result for each time window based on the interference fringes; Performing cluster selection based on the measurement results of the time window, and re-measuring using the selection results to obtain a final measurement result; The modulation of attention to the wavelength of the incident light includes weighting the analysis results of N different wavelengths; Randomly select a wavelength from N interference signals with different wavelengths and set it as m; After obtaining the interference fringe pattern generated by the laser of each wavelength, the interference fringe pattern corresponding to each wavelength is analyzed, and the analysis results of each wavelength are integrated; During integration, the parsing results corresponding to wavelength m are subjected to attention weighting processing; The analytical process includes obtaining the surface morphological characteristics of the film by performing surface fitting on the edge position and morphology of the interference fringes; calculating the thickness information using the spacing of the interference fringes, the thickness of the film, the refractive index, the incident angle, and the wavelength of the light wave; The adjusting of attention to different wavelengths by sliding the time window includes setting a fixed time window length, randomly selecting a wavelength in each time window, and performing attention weighting processing on the corresponding analysis result; The constraints when selecting wavelengths are: in N consecutive time windows, the selected wavelengths are all different; The measurement result of each time window includes weighting the analysis results corresponding to N wavelengths in each time window; assuming that, Represents the weight coefficient. If n corresponds to the analytical result corresponding to the unweighted wavelength, then ; If n corresponds to the analytical result corresponding to the weighted wavelength, then ; M represents the weighting coefficient, and M>2; i∈{C, K, G}, C represents the length of the metal film, K represents the width of the metal film, and G represents the thickness of the metal film; For each parameter i of the metal film, M analytical results corresponding to the weighted wavelength are copied according to the weight coefficient, and the analytical results corresponding to the other N-1 wavelengths form a parameter set as the measurement result; Performing cluster selection based on the measurement results of the time window includes obtaining N consecutive time windows from the current time series, and performing cluster selection for each parameter i of the metal film in any time window; The cluster selection includes clustering the measurement results. If there is only one cluster center, the wavelength of the current window attention weight is determined. For the measurement results of the metal film parameter i, The criterion is satisfied; otherwise, it is judged that the criterion is not satisfied; The re-measurement includes removing k wavelengths that do not meet the criteria from N different wavelengths, and if the number of remaining wavelengths is less than N / 2, randomly selecting k other wavelengths, performing interference fringe analysis and cluster selection on the re-selected wavelengths in a sliding time window until the number of remaining wavelengths after removing the wavelengths that do not meet the criteria is not less than N / 2; obtaining and outputting analysis results for the wavelengths that meet the criteria with respect to the parameter i; The k other wavelengths do not include the removed wavelengths; the reselected wavelengths include the remaining wavelengths that meet the criteria and the newly added k other wavelengths.

2. The method for measuring the size of a metal thin film based on optical measurement according to claim 1, wherein: The multi-wavelength laser includes randomly selecting N laser sources of different wavelengths to form an incident light combination; The selected multi-wavelength laser source is irradiated onto the metal film. The laser beam interacts with the surface. A high-resolution CCD or CMOS camera is used to receive the reflected light and capture the interference images at different wavelengths to obtain the interference fringe pattern generated by each wavelength of laser.

3. The method for measuring the size of a metal film based on optical measurement according to claim 2, wherein: The final measurement result includes cluster selection and re-measurement of the two parameters C and K at the same time; after the output is completed, cluster selection and re-measurement of the parameter G are performed; When all three parameters have completed cluster selection and re-measurement and output, the test process ends; averaging the analytical results of the parameter i corresponding to the wavelengths that meet the criterion to obtain the final measurement result; The process of randomly selecting multiple wavelengths includes the following constraints: N>3; The maximum difference of multiple randomly selected wavelengths is not less than E; Wherein, E represents the threshold value of the maximum wavelength difference.

4. A metal film size measurement system based on optical measurement using the method according to any one of claims 1 to 3, characterized in that: The acquisition unit irradiates the metal film with a multi-wavelength laser to generate an interference signal; The attention control unit modulates the wavelength of the incident light and adjusts the attention of different wavelengths by sliding the time window; a measuring unit, which obtains interference fringes in each time window and obtains a measurement result of each time window based on the interference fringes; The output unit performs cluster selection according to the measurement results of the time window, and remeasures using the selection results until the criterion is met to obtain the final measurement result.

5. A computer device comprising: memory and processor; The memory stores a computer program, wherein the processor implements the steps of the method according to any one of claims 1 to 3 when executing the computer program.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

Images