Thin film thickness measuring method and device based on spectrum confocal

Abstract

The invention provides a thin film thickness measuring method and device based on spectrum confocal. The method is applied to the technical field of high-precision measurement. The method comprises the following steps: respectively acquiring a first reflection spectrum signal and a second reflection spectrum signal aiming at a tested film; determining a thin film upper surface wavelength corresponding to the first peak value in the first reflection spectrum signal, and determining a thin film lower surface wavelength corresponding to the second peak value in the second reflection spectrum signal; calculating the position of the upper surface of the film and the position of the lower surface of the film according to the wavelength of the upper surface of the film, the wavelength of the lower surface of the film and a pre-constructed wavelength-displacement model; and calculating the thickness of the thin film according to the upper surface position and the lower surface position of the thin film. According to the invention, accurate identification and extraction of peak value centers of the first peak and the second peak are realized, and the accuracy of thickness measurement is improved.

Description

Technical Field

[0001] This invention relates to the field of high-precision measurement technology, specifically to a thin film thickness measurement method and device based on spectral confocalization. Background Technology

[0002] Displacement measurement plays a vital role in modern industry and scientific research. To meet diverse measurement requirements, various measurement methods have been proposed, represented by handwritten needle profilometers, electromagnetic sensors, and optical sensing probes. Among these methods, optical probes such as laser triangulation sensors, laser interferometry, and color confocal techniques are widely used in many applications due to their flexibility and non-destructive characteristics.

[0003] Spectral confocal microscopy utilizes a broadband light source to generate continuous focal points along the optical axis, rapidly obtaining the axial position of the sample surface from the focused wavelength in the reflection spectrum. However, in laboratory and practical measurements, this dispersive confocal technique is primarily suitable for transparent or translucent samples. In actual thickness measurements, the characteristic peak intensity of the focused wavelength on the lower surface is often low, resulting in a low local signal-to-noise ratio, affecting the accuracy of peak extraction and thus the precision of thickness measurement.

[0004] Current confocal thickness measurement systems often increase the intensity of characteristic peaks on the lower surface by increasing the power of the light source to achieve focusing on the lower surface of the thin film, or they use multiple measurements to eliminate errors. However, increasing the light power can lead to oversaturation of the light intensity at the focusing wavelength on the upper surface, affecting the accuracy of peak extraction and also posing a risk of exceeding the measurement range of the spectrometer, making measurement impossible. Furthermore, replacing the spectrometer with a compatible one would significantly increase the thickness measurement cost. In addition, increasing the number of measurements reduces measurement efficiency. Summary of the Invention

[0005] In view of this, the present invention provides a thin film thickness measurement method based on spectral confocal microscopy, comprising: A first reflection spectrum signal and a second reflection spectrum signal are collected for the thin film under test, wherein the first reflection spectrum signal has a first peak corresponding to the upper surface of the thin film, and the second reflection spectrum signal has a second peak corresponding to the lower surface of the thin film; The upper surface wavelength of the thin film corresponding to the first peak value is determined in the first reflection spectrum signal, and the lower surface wavelength of the thin film corresponding to the second peak value is determined in the second reflection spectrum signal; The positions of the upper and lower surfaces of the thin film are calculated based on the wavelength of the upper surface, the wavelength of the lower surface, and a pre-built wavelength-displacement model. The film thickness is calculated based on the position of the upper surface of the film and the position of the lower surface of the film.

[0006] Optionally, a first reflectance spectral signal and a second reflectance spectral signal are acquired for the thin film under test, including: A light beam of different intensities is emitted onto a thin film using a light source, and the reflected light focused on the surface of the thin film is collected by a spectrometer to obtain a first reflection spectral signal and a second reflection spectral signal. The first reflection spectral signal is used to extract the first peak corresponding to the upper surface of the thin film, and the second reflection spectral signal is used to extract the second peak corresponding to the lower surface of the thin film.

[0007] Optionally, a first reflectance spectral signal and a second reflectance spectral signal are acquired for the thin film under test, including: The reflected light focused on the surface of the thin film is collected by a spectrometer at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal. The length of the first integration time is suitable for identifying a first peak corresponding to the upper surface of the thin film from the first reflection spectral signal, and the length of the second integration time is suitable for identifying a second peak corresponding to the lower surface of the thin film from the second reflection spectral signal.

[0008] Optionally, the length of the second integration time is greater than the length of the first integration time.

[0009] Optionally, a spectrometer is used to collect reflected light focused on the thin film surface at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal, including: The first reflection spectral signal is obtained by using a first spectrometer to collect the reflected light focused on the surface of the thin film at the first integration time; The reflected light focused on the thin film surface is collected using a second spectrometer at a second integration time to obtain the second reflection spectral signal.

[0010] Optionally, a spectrometer is used to collect reflected light focused on the thin film surface at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal, including: The first reflection spectral signal is obtained by using the first channel of the spectrometer to collect the reflected light focused on the surface of the thin film at the first integration time; The second reflection spectral signal is obtained by using the second channel of the spectrometer to collect the reflected light focused on the thin film surface at the second integration time.

[0011] Optionally, a spectrometer is used to collect reflected light focused on the thin film surface at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal, including: The first reflection spectral signal is obtained by using a spectrometer to collect the reflected light focused on the surface of the thin film at the first integration time; After a preset switching time, the spectrometer is used to collect the reflected light focused on the thin film surface at the second integration time to obtain the second reflection spectral signal.

[0012] Optionally, before determining the upper surface wavelength of the thin film corresponding to the first peak value in the first reflection spectral signal, and before determining the lower surface wavelength of the thin film corresponding to the second peak value in the second reflection spectral signal, the method further includes: The reflected light without background in the state of the thin film under test was collected to obtain the first background spectral signal and the second background spectral signal. Remove the first background spectral signal and the second background spectral signal from the first reflection spectral signal and the second reflection spectral signal, respectively.

[0013] Optionally, a spectrometer is used to collect reflected light focused on the thin film surface at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal, including: The reflected light focused on the thin film surface is collected multiple times using a spectrometer at the first integration time until the integration time reaches the second integration time, thus obtaining multiple first reflection spectral signals. The reflected light focused on the thin film surface is collected using a spectrometer at the second integration time to obtain the second reflection spectral signal.

[0014] Optionally, determining the wavelength of the upper surface of the thin film corresponding to the first peak value in the first reflection spectral signal and determining the wavelength of the lower surface of the thin film corresponding to the second peak value in the second reflection spectral signal includes: Multiple thin film surface wavelengths corresponding to the first peak value are determined from multiple first reflection spectral signals, and the multiple thin film surface wavelengths are averaged to obtain the actual thin film surface wavelength. The lower surface wavelength of the thin film corresponding to the second peak is determined in the second reflection spectrum signal.

[0015] Optionally, determining the wavelength of the upper surface of the thin film corresponding to the first peak value in the first reflection spectral signal and determining the wavelength of the lower surface of the thin film corresponding to the second peak value in the second reflection spectral signal includes: The wavelength of the thin film surface corresponding to the first peak value is determined from the first reflection spectrum signal acquired in the last acquisition; The lower surface wavelength of the thin film corresponding to the second peak is determined in the second reflection spectrum signal.

[0016] Alternatively, the film thickness can be calculated using the following method: , , in, For film thickness, The angle of reflection is the exit angle of the reflected light. The refractive index of the medium on the thin film surface, Let be the refractive index of the thin film. The wavelength of the lower surface of the thin film is [wavelength]. The wavelength is the wavelength of the upper surface of the thin film. , This is a preset function.

[0017] A second aspect of the present invention provides a method for measuring the thickness of multilayer thin films based on spectral confocal microscopy, characterized in that it includes: The thickness of each layer in a multilayer film is measured layer by layer according to the above-described film thickness measurement method. The thickness of the multilayer film is obtained based on the thickness of each layer.

[0018] A third aspect of the present invention provides a thin film thickness measurement device based on spectral confocal microscopy, the device comprising: a processor and a memory connected to the processor; wherein the memory stores instructions executable by the processor, the instructions being executed by the processor to cause the processor to perform the aforementioned thin film thickness measurement method based on spectral confocal microscopy.

[0019] A fourth aspect of the present invention provides a multilayer thin film thickness measurement device based on spectral confocal microscopy. The device includes: a processor and a memory connected to the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor to cause the processor to perform the above-described multilayer thin film thickness measurement method based on spectral confocal microscopy.

[0020] This embodiment precisely captures the reflectance spectral signal of the thin film surface by varying the light intensity of the light source and using a spectrometer with different integration times. For the acquisition of the reflectance spectral signal from the upper surface of the thin film, the light intensity of the light source is adjusted appropriately, or the integration time of the spectrometer is set appropriately to effectively control noise in the spectral signal, ensuring a high signal-to-noise ratio for the signal from the upper surface and making the reflectance spectral signal characteristics clear. For the acquisition of the emission spectral signal from the lower surface of the thin film, considering the energy loss when light penetrates the film, the light intensity of the light source is increased to enhance the intensity of the reflectance spectrum, or the integration time is adjusted to allow the weak reflectance spectral signal from the lower surface to accumulate sufficiently, thereby improving the signal-to-noise ratio of the corresponding lower surface reflectance spectral signal. Then, by extracting the peak values ​​of the two reflectance spectral signals, the wavelengths corresponding to the upper and lower surfaces of the thin film are determined. Combined with a pre-constructed wavelength-displacement correspondence model, the positions of the upper and lower surfaces of the thin film can be accurately calculated, and the thickness of the thin film can then be calculated based on these positional information. This technical solution improves the stability of peak extraction through time-division control. By setting the light intensity of the light source or optimizing the integration time, it achieves accurate identification and extraction of the peak centers of the first and second peaks, thereby significantly improving the testing accuracy and stability of the measurement system and enhancing the accuracy of thickness measurement. Attached Figure Description

[0021] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a reflectance spectrum diagram from an embodiment of the present invention; Figure 2 This is a magnified view of a peak in an embodiment of the present invention; Figure 3 This is a magnified view of a two-peak area in an embodiment of the present invention; Figure 4 This is a flowchart of a thin film thickness measurement method based on spectral confocalization in an embodiment of the present invention; Figure 5 These are the reflectance spectra at two integration times in an embodiment of the present invention; Figure 6 This is the spectral diagram of reflectance spectrum 1 in this embodiment of the invention; Figure 7 This is the spectral diagram of reflectance spectrum 2 in this embodiment of the invention. Detailed Implementation

[0023] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0025] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can also refer to the internal connection of two components; and they can refer to a wireless connection or a wired connection. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0026] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0027] like Figure 1 As shown, the reflectance spectral characteristics of the wavelengths focused on the upper and lower surfaces of a silicon carbide thin film, measured using a spectral confocal system at a certain integration time, include two characteristic peaks. The center of the peaks where the light intensity of the two characteristic peaks is the focused wavelength of the upper and lower surfaces. The specific locations of the focused wavelengths on the upper and lower surfaces can be found using filtering and peak extraction algorithms. In the following spectra, the vertical axis represents light intensity, and the horizontal axis represents wavelength. After receiving the reflectance spectral signal, light source characteristics are removed and normalized to 0-1 to improve the system's processing capability.

[0028] according to Figure 2 A magnified view of the peak value and Figure 3The magnified view of the peak values ​​of the two peaks shows that the peak has a large peak intensity, a tall and narrow characteristic peak, and a small half-width at half maximum (FWHM). The peak center extraction is relatively stable and has high resolution. However, the second peak, i.e. the wavelength at the lower surface focus, is affected by factors such as refraction and material absorption. The received reflected light intensity is low, the FWHM is large, the data quality is poor, the signal-to-noise ratio is too low, and there are many local outliers. In this case, the peak finding algorithm performs poorly and has low resolution.

[0029] The thickness compensation of the spectral confocal displacement sensor is mainly affected by the two-peak focusing wavelength, and the focusing position is a parameter used to calculate the thickness result. The angle of reflection of the light at the corresponding wavelength and corresponding wavelength refractive index Both are affected by the wavelength of the lower surface of the thin film. Changes have an impact. Therefore, the stability and accuracy of the two peaks directly affect the stability and accuracy of thickness measurement.

[0030] Therefore, as Figure 4 As shown, this embodiment of the invention provides a thin film thickness measurement method based on spectral confocal microscopy. This method is executed by an electronic device such as a computer or server, and specifically includes: S1, acquire a first reflection spectrum signal and a second reflection spectrum signal for the thin film under test, wherein the first reflection spectrum signal has a first peak corresponding to the upper surface of the thin film, and the second reflection spectrum signal has a second peak corresponding to the lower surface of the thin film.

[0031] This can be understood as follows: If only the first peak corresponding to the upper surface of the thin film is extracted using the first reflection spectrum, then the signal-to-noise ratio (SNR) of the first peak corresponding to the upper surface of the thin film in the first reflection spectrum signal to be acquired is very high, while the SNR of the second peak corresponding to the lower surface of the thin film is extremely low due to refraction and other reasons. Conversely, if only the second peak corresponding to the lower surface of the thin film is extracted using the second reflection spectrum, then the first peak corresponding to the upper surface of the thin film in the first reflection spectrum signal to be acquired is in an oversaturated state, and this peak is distorted, so there is no need to extract information from it. However, the SNR of the second peak corresponding to the lower surface of the thin film in the second reflection spectrum signal is significantly improved, and the second peak can be clearly extracted. Therefore, there are at least three implementation methods: The first method: S1a, using a light source to emit light beams of different intensities onto the thin film, and using a spectrometer to collect the reflected light focused on the surface of the thin film, a first reflection spectral signal and a second reflection spectral signal are obtained, wherein the first reflection spectral signal is used to extract the first peak corresponding to the upper surface of the thin film, and the second reflection spectral signal is used to extract the second peak corresponding to the lower surface of the thin film.

[0032] Specifically, the signal-to-noise ratio of the two peaks can be improved by adjusting the light intensity of the light source, and modulated light intensity can be used.

[0033] For example, the modulated light intensity can be adjusted by using a modulated light source. The modulated light source can achieve a relatively uniform distribution in the white light band of 400-700nm, and the intensity of the longer wavelength light is greater than that of the shorter wavelength light. The specific implementation scheme is as follows: Multi-channel LED combination system: This system combines monochromatic LEDs covering different wavelengths (e.g., blue, cyan, green, yellow, red), and independently adjusts the drive current of each channel to dynamically control the spectral distribution. Specifically, LEDs covering 400-700nm are first selected (e.g., 400-450nm (blue), 450-500nm (cyan), 500-570nm (green), 570-600nm (yellow), 600-700nm (red)). A multi-channel constant current driver is used to independently adjust the intensity of each LED via PWM (Pulse Width Modulation) or analog dimming, allowing for a larger drive current and higher light intensity in the longer wavelength (red) channels. Then, an integrating sphere, diffuser, or fiber optic mixer is used to uniformly mix the light of different wavelengths. Finally, a spectrometer monitors the output in real time and provides feedback to adjust the intensity of each channel, ensuring spectral uniformity (intensity difference between wavelengths <20%) and higher integrated energy in the longer wavelength bands. This solution features fast modulation speed (LED response time <1μs), flexible adjustment of spectral shape, and controllable cost, making it suitable for laboratory or industrial applications.

[0034] Broadband Light Source + Dynamic Filtering System: This system utilizes halogen lamps or white LEDs as the base light source, combined with tunable filters (such as Liquid Crystal Tunable Filters (LCTFs)) to selectively enhance transmittance in the long wavelength range. Specifically, a halogen lamp (continuous spectrum 400-2500nm) or a high color rendering white LED is selected as the base light source. Then, an LCTF or an Acousto-Optic Tunable Filter (AOTF) is used to dynamically adjust the transmission wavelength, enhancing transmittance in the 600-700nm region. Finally, the intensity ratio of different wavelengths is adjusted by filter modulation or by adding a light intensity modulator (such as a Digital Micromirror Device (DMD)). This solution provides a continuous, gapless spectrum with superior uniformity, making it suitable for scenarios requiring high spectral resolution.

[0035] A fluorescence conversion + red light enhancement scheme: This scheme uses a blue LED to excite a yellow phosphor to produce white light, supplemented by a red LED to compensate for the intensity in the long wavelength range. Specifically, a high-brightness blue LED (e.g., 450nm) is first used to excite YAG (Yttrium Aluminum Garnet) phosphor to generate a broad-spectrum white light in the 400-650nm range. Then, a high-power red LED (e.g., 660nm) is integrated and independently driven to enhance the long wavelength region. Finally, the current ratio between the blue and red LEDs is adjusted to significantly improve the overall spectral intensity in the 600-700nm region. This scheme has a compact structure, and the red light intensity can be directly controlled by the driving current.

[0036] The above solutions use a spectrometer to measure the light intensity at each wavelength and optimize the driving parameters using algorithms (such as the least squares method) to balance short-wavelength uniformity and long-wavelength enhancement. Long-wavelength LEDs (such as red light) are typically less efficient, requiring enhanced heat dissipation design (such as heat sinks and fans). Digital interfaces (such as USB and WiFi) are provided to support user-defined spectral curves, for example, setting the intensity of the red light region to 1.5-2 times that of blue light. Overall, a multi-channel LED combination system is recommended. If the budget allows and a continuous spectrum is required, a dynamic filtering scheme can be considered.

[0037] The second method: S1b, using a spectrometer, reflected light focused on the surface of the thin film is collected at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal, wherein the length of the first integration time is suitable for identifying a first peak corresponding to the upper surface of the thin film from the first reflection spectral signal, and the length of the second integration time is suitable for identifying a second peak corresponding to the lower surface of the thin film from the second reflection spectral signal.

[0038] The characteristics of the reflectance spectral signal are related to the integration time. The value of the first integration time should ensure that the first peak in the first reflectance spectrum does not exceed the maximum range, and that it has a center of symmetry, no missing vertices, and a clear peak value. This means that at this integration time, the first reflectance spectrum should be clearly visible and not affected by oversaturation or signal loss. The purpose of the second integration time is to make the second peak in the second spectrum more obvious, without having to consider whether the first peak is saturated (out of range). This means that the choice of the second integration time may lead to the saturation of the first peak value, but this is permissible and acceptable for extracting the second peak value.

[0039] The third method: When acquiring reflectance spectral signals using a spectrometer, signal strength and signal-to-noise ratio (SNR) are two key factors affecting the accuracy and reliability of peak measurement. Signal strength directly affects the clarity of the target signal; a weak signal may lead to information loss, while an overly strong signal can cause saturation, distorting the peak measurement results. Simultaneously, the SNR reflects the ratio of useful information to noise in the signal; a high SNR indicates a cleaner and more reliable signal. To achieve effective control of signal strength and optimization of the SNR, this scheme employs both a signal strength model and a SNR model.

[0040] For example, signal strength model: light source power Driven by current The relationship between voltage U and voltage is as follows: , in, This indicates the photoelectric conversion efficiency of the light source (unit: W / A×V, which needs to be calibrated experimentally; the photoelectric conversion efficiency of a typical LED light source is between 20-30%).

[0041] Signal strength output by the spectrometer This reflects the magnitude of the actual collected reflectance spectral signal intensity, and the relationship is as follows: , in, It represents the fiber optic transmission efficiency (including coupling rate and wavelength-dependent loss, etc., and is dimensionless). The reflectance of the lower surface of the thin film is dimensionless. Indicates the integration time (in seconds). The detector response coefficient (unit: count / W×s, requires experimental calibration), i.e., the unit signal strength of the spectrometer. The corresponding signal strength, Indicates the upper limit of the spectrometer's dynamic range, such as 16 bits corresponding to... .

[0042] As can be seen from the above formula, increasing signal strength can be achieved by increasing the power of the light source. and increase integration time This is done in various ways. Increasing the light source power can increase the driving current. Alternatively, the voltage U can be adjusted within a certain range. Integral time. The longer the duration, the more accumulated reflectance spectral signal the spectrometer acquires. With other parameters remaining constant, the signal intensity... It will increase accordingly. At the same time, when near The signal may become oversaturated, leading to a decrease in the nonlinearity ratio and signal-to-noise ratio. Therefore, to ensure measurement accuracy and reliability, the signal strength must meet certain constraints, such as... Target signal strength That is, the second peak intensity on the lower surface of the film needs to satisfy If the current signal strength It can be adjusted as follows: Adjust only the drive current: , in, This indicates the adjusted drive current. This indicates the driving current of the current signal. This indicates the maximum allowable current of the light source.

[0043] Adjusting only the points time: , in, This indicates the adjusted integration time. This represents the integration time of the currently acquired signal. This indicates the maximum integration time allowed by the system.

[0044] Joint adjustment of drive current and integral time: If a single parameter is limited (e.g.) or (This requires joint adjustment based on output characteristics) , And satisfy .

[0045] For example, the signal-to-noise ratio model: when the driving current and integration time are adjusted according to the signal strength model, the second peak intensity on the lower surface of the thin film satisfies... Subsequently, signal quality becomes a key factor affecting the accuracy of peak measurement, and signal-to-noise ratio (SNR) is a crucial indicator for measuring signal quality. Even if the signal strength meets the requirements, if the SNR is too low, the peak measurement result may still be inaccurate. Therefore, the SNR model relationship is as follows: , in, Indicates the signal-to-noise ratio. This represents shot noise (as opposed to signal strength). (proportional to the square) This indicates the detector noise of the spectrometer. , Indicates readout noise. Indicates dark current noise. This indicates background noise.

[0046] As can be seen from the signal-to-noise ratio (SNR) model, signal strength can be increased to improve the SNR. Approaching the upper limit At the same time, the detector noise of the spectrometer can be reduced through the cooling system. Reduce background noise by using optical path shielding or filters. This improves the accuracy of peak measurement.

[0047] S2, determine the wavelength of the upper surface of the thin film corresponding to the first peak in the first reflection spectrum signal, and determine the wavelength of the lower surface of the thin film corresponding to the second peak in the second reflection spectrum signal.

[0048] S3, calculate the positions of the upper and lower surfaces of the thin film based on the wavelengths of the upper and lower surfaces of the thin film and the pre-built wavelength-displacement model; S4, calculate the film thickness based on the positions of the upper and lower surfaces of the film.

[0049] This embodiment precisely captures the reflectance spectral signal of the thin film surface by varying the light intensity of the light source and using a spectrometer with different integration times. For the acquisition of the reflectance spectral signal from the upper surface of the thin film, the light intensity of the light source is adjusted appropriately, or the integration time of the spectrometer is set appropriately to effectively control noise in the spectral signal, ensuring a high signal-to-noise ratio for the signal from the upper surface and making the reflectance spectral signal characteristics clear. For the acquisition of the emission spectral signal from the lower surface of the thin film, considering the energy loss when light penetrates the film, the light intensity of the light source is increased to enhance the intensity of the reflectance spectrum, or the integration time is adjusted to allow the weak reflectance spectral signal from the lower surface to accumulate sufficiently, thereby improving the signal-to-noise ratio of the corresponding lower surface reflectance spectral signal. Then, by extracting the peak values ​​of the two reflectance spectral signals, the wavelengths corresponding to the upper and lower surfaces of the thin film are determined. Combined with a pre-constructed wavelength-displacement correspondence model, the positions of the upper and lower surfaces of the thin film can be accurately calculated, and the thickness of the thin film can then be calculated based on these positional information. This technical solution improves the stability of peak extraction through time-division control. By setting the light intensity of the light source or optimizing the integration time, it achieves accurate identification and extraction of the peak centers of the first and second peaks, thereby significantly improving the testing accuracy and stability of the measurement system and enhancing the accuracy of thickness measurement.

[0050] Further, step S4 calculates the film thickness using the following formula: , , in, For film thickness, The angle of reflection is the exit angle of the reflected light. The refractive index of the medium on the thin film surface, Let be the refractive index of the thin film. The wavelength is at the lower surface of the thin film. The wavelength is the wavelength of the upper surface of the thin film. , This is a preset function.

[0051] The preset function is obtained by calibration based on measured data, specifically as follows: , in, This indicates the parameter corresponding to the center of the spectral peak. Location, These parameters are determined using the parameters corresponding to the spectral peak center and the distance. Through the aforementioned step measurement process, the parameters corresponding to the spectral peak center and the reliable distance values ​​at each step position can be collected. Using this large amount of data, the [parameter / value] can be calculated. The value of is taken to obtain the function. The wavelength of the lower surface of the thin film is then determined. and the wavelength of the thin film surface By substituting into the function, we can obtain and This gives us the positions of the upper and lower surfaces of the film.

[0052] In one embodiment, the length of the second integration time is greater than the length of the first integration time. A longer second integration time ensures that a clear second peak is extracted.

[0053] Two spectrometers are set up in the reflected light path, simultaneously receiving the reflected spectral signals from the dispersive frequency-selective system. For example, one spectrometer is set with an integration time of 10 ms, and the other with an integration time of 50 ms, denoted as spectrometers A and B. It is foreseeable that the spectral signal from the first spectrometer A during the 10 ms integration time will have two peaks, one focused on the upper surface of the thin film and the other on the lower surface. The wavelengths of these two peaks can be extracted. The other spectrometer, B, with a 50 ms integration time, will experience saturation of the relative intensity of one peak due to the longer integration time; therefore, the wavelengths of the two peaks will be primarily extracted.

[0054] Two spectrometers are used to simultaneously receive the reflection spectral signals from a dispersive frequency-selective system along the reflected light path. The integration times of the two spectrometers are different. For the same object and the same location, the reflected light spectrum changes after varying the integration time, as shown below. Figure 5 As shown, the horizontal axis represents wavelength, and the vertical axis represents signal intensity. It is evident that the first peak gradually saturates, while the FWHM of the second peak decreases, improving resolution and thus enhancing the stability of the second peak. Therefore, using a short integration time to extract the peak value of the first peak and increasing the integration time to extract the peak value of the second peak improves the stability and accuracy of the measurement of the positions on the upper and lower surfaces of the thin film.

[0055] The following are some embodiments for using a spectrometer to collect reflected light focused on the surface of a thin film at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal: First embodiment: S1b11, using a first spectrometer to collect reflected light focused on the surface of the thin film at a first integration time, a first reflection spectral signal is obtained; S1b12, using a second spectrometer to collect the reflected light focused on the thin film surface at the second integration time, obtains the second reflection spectral signal.

[0056] In this embodiment, the length of the first integration time is less than the length of the second integration time. Alternatively, the first integration time can be equal to the second integration time, and the range of the first spectrometer can be greater than the range of the second spectrometer.

[0057] For example, using two spectrometers with different measurement ranges and the same integration time (e.g., 10 ms), simultaneously acquiring reflected light focused on the thin film surface, we obtain a first reflection spectral signal and a second reflection spectral signal. At this integration time, the first reflection spectral signal acquired by the larger-range spectrometer A has two peaks: one at the upper surface of the thin film and the other at the lower surface. In this case, only the first peak is extracted. However, in the smaller-range spectrometer B, the first peak of the second reflection spectral signal at the same integration time is saturated and cannot be displayed, while the second peak has a distinct characteristic, allowing for the extraction of its position. By processing the first peak data acquired by spectrometer A and the second peak data acquired by spectrometer B, respectively, the current thin film thickness value can be obtained.

[0058] Second embodiment: S1b21, using the first channel of the spectrometer to collect the reflected light focused on the thin film surface at the first integration time, the first reflection spectral signal is obtained; S1b22, using the second channel of the spectrometer to collect the reflected light focused on the thin film surface at the second integration time, obtains the second reflection spectral signal.

[0059] Considering the complexity and cost of the two spectrometers, a multi-channel spectrometer can be used instead. Two channels are used to receive the reflection spectrum at different integration times. The subsequent processing and thickness calculation are the same as the steps for the two spectrometers mentioned above, and will not be repeated here.

[0060] Third embodiment: S1b31, the first reflection spectral signal is obtained by using a spectrometer to collect the reflected light focused on the surface of the thin film at the first integration time; S1b32, after a preset switching time, uses a spectrometer to collect the reflected light focused on the thin film surface at the second integration time to obtain the second reflection spectral signal.

[0061] This embodiment uses a spectrometer, which is debugged and calibrated before the measurement begins, and an appropriate integration time is selected to extract the first and second peak wavelengths, such as 10ms and 50ms.

[0062] The entire acquisition time is set to 70ms. First, the spectrometer acquires data over 10ms, using a multi-threaded task for peak extraction. Then, it switches to a 50ms integration time, with a switching time set to 10ms. After the switch stabilizes, it acquires reflectance spectral data over a 50ms integration time, again using a multi-threaded task for peak extraction. This method enables the use of a single spectrometer to acquire spectral data in a single session and extract one or two peaks separately, achieving a system response time of 100ms.

[0063] Fourth embodiment: S1b41, using a spectrometer, the reflected light focused on the thin film surface is collected multiple times at the first integration time until the integration time reaches the second integration time, thus obtaining multiple first reflection spectral signals; S1b42 uses a spectrometer to collect reflected light focused on the thin film surface at the second integration time to obtain the second reflection spectral signal.

[0064] This embodiment is applicable when the thickness of the film being tested is constantly changing during dynamic processing, such as when the film thickness decreases during polishing. Since the length of the second integration time is longer than the length of the first integration time, a certain period has passed by the time the position of the lower surface of the film is obtained through the acquired second spectral signal. Therefore, the previously acquired second spectral signal is no longer accurate in determining the position of the lower surface of the film. Thus, taking two spectrometers as an example, while the second spectrometer receives reflected light, the first spectrometer also simultaneously receives reflected light continuously for the first integration time. In this way, by the time the second integration time is reached, the first spectrometer will receive multiple first reflection spectral signals.

[0065] Further, in step S2, determining the wavelength of the upper surface of the thin film corresponding to the first peak value in the first reflection spectral signal and determining the wavelength of the lower surface of the thin film corresponding to the second peak value in the second reflection spectral signal specifically includes: S21a, determine multiple thin film surface wavelengths corresponding to the first peak value from multiple first reflection spectrum signals, and average the multiple thin film surface wavelengths to obtain the actual thin film surface wavelength. S22a, determine the wavelength of the lower surface of the thin film corresponding to the second peak in the second reflection spectrum signal.

[0066] Furthermore, in step S2, determining the wavelength of the upper surface of the thin film corresponding to the first peak value in the first reflection spectral signal and determining the wavelength of the lower surface of the thin film corresponding to the second peak value in the second reflection spectral signal can also be: S21b, determine the wavelength of the upper surface of the thin film corresponding to the first peak in the first reflection spectrum signal acquired in the last acquisition; S22b, determine the wavelength of the lower surface of the thin film corresponding to the second peak in the second reflection spectrum signal.

[0067] In scenarios where the thickness of the film being measured changes continuously during dynamic processing, this embodiment obtains multiple first reflection spectral signals by repeatedly acquiring reflected light focused on the film surface over a first integration time, and obtains a second reflection spectral signal over a second integration time. There are two effective methods for determining the wavelengths of the upper and lower surfaces of the film. The first method (S21a and S22a) averages the wavelengths of the upper surface of the film corresponding to the first peak in the multiple first reflection spectral signals to obtain the actual upper surface wavelength, and simultaneously determines the lower surface wavelength corresponding to the second peak in the second reflection spectral signal. This method effectively reduces measurement errors, and the averaging process makes the upper surface wavelength closer to the true value. The second method (S21b and S22b) determines the upper surface wavelength from the last acquired first reflection spectral signal and the lower surface wavelength from the second reflection spectral signal. This method can reflect the latest state of the upper surface of the film during dynamic processing in a timely manner, and combined with the lower surface wavelength, it enables more accurate and real-time measurement of the dynamically changing film thickness.

[0068] Optionally, after determining the corresponding upper surface wavelength and lower surface wavelength of the thin film using the first reflection spectral signal and the second reflection spectral signal obtained in the first embodiment, the two-peak data extracted by the second spectrometer can be verified using the two-peak data of the first spectrometer, for example: The dual-peak data from spectrometer A is used to verify the dual-peak data extracted by spectrometer B. Since the integration time of spectrometer A is 10ms, within 50ms of spectrometer B acquiring reflected light information once, A can collect and output the peak wavelengths of the dual peaks five times. The average value of the focused wavelength is calculated and used as a reference. The difference between the peak wavelength extracted by spectrometer B and the reference data is judged. If it is less than a certain threshold, it is considered as a same-frequency selection, and the lower surface wavelength data obtained by the second spectrometer is determined to be correct. If it is greater than the threshold, it proves that there is a certain deviation in the lower surface focused wavelengths collected by the two spectrometers, and the data of that time can be discarded.

[0069] In one embodiment, prior to step S2, the method further includes: The reflected light without background in the state of the thin film under test was collected to obtain the first background spectral signal and the second background spectral signal. Remove the first background spectral signal and the second background spectral signal from the first reflection spectral signal and the second reflection spectral signal, respectively.

[0070] This embodiment effectively eliminates the interference of ambient light, instrument noise, and other background spectral signals on the reflectance spectral signals by removing background spectral signals. This makes the first and second reflectance spectral signals used for subsequent analysis more accurate, laying a solid foundation for subsequent operations such as accurately determining the corresponding wavelengths on the upper and lower surfaces of the thin film and accurately calculating the film thickness. This improves the reliability and accuracy of the entire measurement process and results.

[0071] Taking the integration time as an example, the experimental measurement is carried out according to the above operation procedure S1-S4.

[0072] First, a Y-type optical fiber is used to connect the light source to a spectrometer, with one end connected to a lens. The integration time is adjusted to 8ms, and the background light (ALL_back_8ms) in the absence of the analyte is recorded. During the measurement, the acquired spectral data will be subtracted from ALL_back_8ms in real time.

[0073] Record the reflected light spectrum at the current location, denoted as reflection spectrum 1. Extract the upper and lower surface focused peaks from reflection spectrum 1, and calculate the peak extraction standard deviation. The processed data is as follows: Figure 6 As shown, the horizontal axis represents wavelength, and the vertical axis represents resolution. The upper surface focuses on a wavelength of 557.20 nm, and the lower surface focuses on a wavelength of 579.53 nm.

[0074] Optionally, under static laboratory conditions, multiple measurements can be taken to verify the repeatability of peak extraction.

[0075] The standard deviation is calculated using the Bessel formula. , in, It is the arithmetic mean of n measurements.

[0076] , Alternatively, the mean absolute error (MAD) can be used to evaluate the stability of multiple measurements.

[0077] Fifty data acquisitions were performed at the same location. The calculated peak repeatability std = 0.00867 nm for the upper surface and 0.0374 nm for the lower surface. It can be seen that the stability of the lower surface is significantly worse than that of the upper surface.

[0078] Then, the integration time was adjusted to 50ms, and the reflected light data was received using a second spectrometer. The background light (ALL_back_50ms) was recorded when there was no analyte. The spectral data acquired during the measurement will be subtracted from ALL_back_50ms.

[0079] Record the reflection spectrum at the current location, denoted as reflection spectrum 2. At this point, one peak in the reflection spectrum is in a saturated state; extract the focusing wavelength from the lower surface. Extract the effective data segment and use the algorithm to extract the peak. The result after data processing is as follows: Figure 7 As shown, the horizontal axis represents wavelength, and the vertical axis represents resolution.

[0080] The stability was analyzed by comparing multiple measurement experiments. The standard deviation of the peak extracted at the same position was calculated. The stability of the two peaks was 0.01834nm, which is significantly improved. Therefore, the two peak data with the smaller standard deviation, that is, the data extracted at the 50ms integration time, was selected for decoding.

[0081] The obtained wavelengths from the upper and lower surfaces are used for decoding, data calculation, and thickness calculation to obtain the final thickness.

[0082] The following are the results of stability and thickness for two-peak extraction at different integration times.

[0083]

[0084] As can be seen from the data in the table above, the stability of the two peaks and the stability of the thickness results improve with the adjustment of the integration time. Furthermore, the thickness of this sample was measured using the Oncon Vision SPI1300H thin film topography inspection equipment and the Nikon-DIGIMICRO digital height gauge (measurement accuracy 1µm, reading accuracy 0.01µm), yielding a result of 361.17µm. The thickness results at different integration times are all within the allowable error range, meeting the measurement requirements.

[0085] In one embodiment, the present invention also provides a method for measuring the thickness of multilayer thin films based on spectral confocal microscopy. This method is executed by an electronic device such as a computer or server, and specifically includes: The thickness of each layer in a multilayer film is measured layer by layer according to the above-described film thickness measurement method. The thickness of the multilayer film is obtained based on the thickness of each layer.

[0086] Specifically, for multilayer thin films, the thickness of each layer is measured starting from the top layer, and then the sum of the thicknesses of each single layer is calculated to obtain the thickness of the multilayer film. For example, assuming there are two layers, the thickness of the first layer is calculated using the formula for calculating the thickness of a single layer, then the thickness of the second layer is calculated, and the sum of the two thicknesses gives the total thickness of the two-layer film.

[0087] The single-layer and multi-layer film thickness measurement methods in all the above embodiments are applicable to static processing thickness measurement. If used for dynamic processing thickness measurement, the wavelength of the upper surface and the wavelength of the lower surface of the film can be determined according to the following embodiments: determine whether the change in film thickness during the integration time is less than the minimum thickness resolution of the spectrometer. If the change in film thickness during the integration time is less than the minimum thickness resolution of the spectrometer, then the method of steps S1b41-S1b42 can be executed, and then step S2 can be executed.

[0088] In this embodiment, when the change in film thickness within the integration time is less than the minimum thickness resolution of the spectrometer, the spectrometer cannot detect subtle thickness changes in the film during processing. This can lead to a significant deviation between the acquired film thickness data and the actual film thickness. This solution employs an alternating integration time method, acquiring the first reflectance spectral signal multiple times. Two specific data processing methods are used: first, averaging the wavelength data from multiple film surface measurements corresponding to the first peak; second, selecting only the wavelength data from the last measurement. This effectively reduces the error caused by the spectrometer's inability to distinguish subtle thickness changes, thereby improving the accuracy of film thickness measurement.

[0089] In another embodiment, if the change in film thickness during the integration time is greater than the minimum thickness resolution of the spectrometer, the spectrometer is used to collect the reflected light focused on the film surface during the second integration time to obtain the target reflection spectrum signal. The half-peak width algorithm is used to determine the wavelength of the upper surface of the thin film corresponding to the first peak value in the target reflectance spectrum signal, and to determine the wavelength of the lower surface of the thin film corresponding to the second peak value in the target reflectance spectrum signal.

[0090] In this implementation, when the change in film thickness during the integration time exceeds the minimum thickness resolution of the spectrometer, the spectrometer can respond to the change in film thickness. Because the thickness change is too large, if the two integration times of the above embodiment are used, the film thickness will have changed during the interval between the two integration times, leading to a significant deviation between the final calculated film thickness and the actual thickness. Therefore, this solution uses only one large integration time, i.e., the second integration time is used to collect the reflected light focused on the film surface to obtain the target reflectance spectral signal. However, this method will cause oversaturation of the first peak data in the target reflectance spectral signal, resulting in the first peak value not being displayed correctly. To address this problem, this solution uses a half-peak width algorithm to determine the wavelength of the upper surface of the film, while the second peak value can be directly obtained from the target reflectance spectral signal. The half-peak width algorithm includes: Based on multiple light intensity lines in the target reflection spectrum signal, determine the wavelength corresponding to half the intensity value of each light intensity line, and calculate multiple symmetrical peak centers based on multiple wavelengths; calculate the average of multiple symmetrical peak centers to obtain the first peak value.

[0091] For example, the first peak value is calculated using the following method. : , , in, Indicates multiple symmetrical peak centers, , This represents the two wavelengths corresponding to any given light intensity line.

[0092] This embodiment addresses the situation where the film thickness variation within the integration time exceeds the minimum thickness resolution of the spectrometer. It employs a single second integration time to collect reflected light, avoiding measurement errors caused by using different thickness data for the same thickness calculation due to the time difference between the two integrations. Simultaneously, the half-width at half-maximum (WWHM) algorithm solves the problem of oversaturated first peak data failing to display the peak value correctly, accurately determining the wavelength on the upper surface of the film and effectively improving the accuracy of film thickness measurement.

[0093] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0094] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0095] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0096] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0097] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A thin film thickness measurement method based on spectral confocal microscopy, characterized in that, include: A light beam of different intensities is emitted onto a thin film using a light source. The reflected light focused on the surface of the thin film is collected by a spectrometer at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal. The first reflection spectral signal contains a first peak corresponding to the upper surface of the thin film, and the second reflection spectral signal contains a second peak corresponding to the lower surface of the thin film. The length of the first integration time is suitable for identifying the first peak corresponding to the upper surface of the thin film from the first reflection spectral signal, and the length of the second integration time is suitable for identifying the second peak corresponding to the lower surface of the thin film from the second reflection spectral signal. The upper surface wavelength of the thin film corresponding to the first peak value is determined in the first reflection spectrum signal, and the lower surface wavelength of the thin film corresponding to the second peak value is determined in the second reflection spectrum signal; The positions of the upper and lower surfaces of the thin film are calculated based on the wavelength of the upper surface, the wavelength of the lower surface, and a pre-built wavelength-displacement model. The film thickness is calculated based on the position of the upper surface of the film and the position of the lower surface of the film.

2. The method according to claim 1, characterized in that, The length of the second integration time is greater than the length of the first integration time.

3. The method according to claim 1, characterized in that, The reflected light focused on the thin film surface is collected using a spectrometer at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal, including: The first reflection spectral signal is obtained by using a first spectrometer to collect the reflected light focused on the surface of the thin film at the first integration time; The reflected light focused on the thin film surface is collected using a second spectrometer at a second integration time to obtain the second reflection spectral signal.

4. The method according to claim 2, characterized in that, The reflected light focused on the thin film surface is collected using a spectrometer at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal, including: The first reflection spectral signal is obtained by using the first channel of the spectrometer to collect the reflected light focused on the surface of the thin film at the first integration time; The second reflection spectral signal is obtained by using the second channel of the spectrometer to collect the reflected light focused on the thin film surface at the second integration time.

5. The method according to claim 2, characterized in that, The reflected light focused on the thin film surface is collected using a spectrometer at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal, including: The first reflection spectral signal is obtained by using a spectrometer to collect the reflected light focused on the surface of the thin film at the first integration time; After a preset switching time, the spectrometer is used to collect the reflected light focused on the thin film surface at the second integration time to obtain the second reflection spectral signal.

6. The method according to claim 2, characterized in that, The reflected light focused on the thin film surface is collected using a spectrometer at a first integration time and a second integration time, respectively, to obtain a first reflection spectral signal and a second reflection spectral signal, including: The reflected light focused on the thin film surface is collected multiple times using a spectrometer at the first integration time until the integration time reaches the second integration time, thus obtaining multiple first reflection spectral signals. The reflected light focused on the thin film surface is collected using a spectrometer at the second integration time to obtain the second reflection spectral signal.

7. The method according to claim 6, characterized in that, Determining the upper surface wavelength of the thin film corresponding to the first peak value in the first reflection spectral signal, and determining the lower surface wavelength of the thin film corresponding to the second peak value in the second reflection spectral signal, includes: Multiple thin film surface wavelengths corresponding to the first peak value are determined from multiple first reflection spectral signals, and the multiple thin film surface wavelengths are averaged to obtain the actual thin film surface wavelength. The lower surface wavelength of the thin film corresponding to the second peak is determined in the second reflection spectrum signal.

8. The method according to claim 6, characterized in that, Determining the upper surface wavelength of the thin film corresponding to the first peak value in the first reflection spectral signal, and determining the lower surface wavelength of the thin film corresponding to the second peak value in the second reflection spectral signal, includes: The wavelength of the thin film surface corresponding to the first peak value is determined from the first reflection spectrum signal acquired in the last acquisition; The lower surface wavelength of the thin film corresponding to the second peak is determined in the second reflection spectrum signal.

9. A method for measuring the thickness of multilayer thin films based on spectral confocal microscopy, characterized in that, include: The thin film thickness measurement method according to claims 1-8 measures the thickness of each layer in a multilayer thin film layer by layer. The thickness of the multilayer film is obtained based on the thickness of each layer.

10. A thin film thickness measurement device based on spectral confocal microscopy, characterized in that, include: A processor and a memory connected to the processor; wherein the memory stores instructions executable by the processor, the instructions being executed by the processor to cause the processor to perform the method as described in any one of claims 1-9.

Images