A method and apparatus for detecting the thickness of a film

Abstract

The application discloses a film thickness detection method and device. The film thickness detection method comprises the following steps: acquiring data of reflectivity of a film surface detection light changing with time to form an experimental waveform; establishing a theoretical echo model; the theoretical echo model is calculated by setting and adjusting adjustable parameters to obtain multiple theoretical waveforms of reflectivity of the film surface detection light changing with time; comparing the experimental waveform with the multiple theoretical waveforms to screen out a theoretical echo model with the highest matching degree with the experimental waveform as an optimal theoretical echo model; and calculating the optimal theoretical waveform of the reflectivity of the film surface detection light changing with time according to the optimal theoretical echo model, and calculating the thickness of the film according to the optimal theoretical waveform. The application can improve the precision and speed of measuring the film thickness, and can guarantee the accuracy and reliability of the detection result.

Description

Technical Field

[0001] The present invention relates to the field of thickness detection technology, and in particular to a method and apparatus for detecting the thickness of a thin film. Background Technology

[0002] With the rise of the electronic and nanoscience era, thin film materials have been widely used in semiconductors, optical devices, microelectromechanical systems and other fields.

[0003] Thin film thickness has a significant impact on its performance and application effects; therefore, accurate measurement of thin film thickness has become a key technology in these industries. Traditional film thickness measurement methods (such as optical interferometry and ellipsometric methods), while highly accurate, have limitations when measuring opaque or multilayer films. Photoacoustic film thickness measurement is commonly used for measuring opaque or multilayer films. The core principle of photoacoustic film thickness measurement is to use laser pulses to excite the material, causing thermal expansion and generating sound waves. These sound waves propagate within the material and are detected. By analyzing the propagation characteristics, propagation time, and sound velocity within the material, the film thickness can be deduced. However, due to crosstalk between multilayer films and attenuation effects from multiple propagations, subsequent effective signals may be suppressed by background noise, leading to significant errors in the detected thickness of each layer in the multilayer film. Therefore, existing photoacoustic film thickness measurement techniques cannot guarantee the accuracy and reliability of the measurement results. Summary of the Invention

[0004] This invention provides a method and apparatus for detecting film thickness, which can reduce the error in measuring film thickness, improve the accuracy and speed of film thickness measurement, and ensure the accuracy and reliability of the detection results.

[0005] In a first aspect, embodiments of the present invention provide a method for detecting thin film thickness, the method comprising:

[0006] Data on the reflectivity of probe light on the thin film surface as a function of time were obtained to form experimental waveforms;

[0007] A theoretical echo model is established, which calculates the theoretical waveforms of the reflectivity of the probe light on the thin film surface as a function of time by setting and adjusting adjustable parameters.

[0008] The experimental waveform is compared with multiple theoretical waveforms, and the theoretical echo model that matches the experimental waveform the most is selected as the optimal theoretical echo model.

[0009] The optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time is calculated based on the optimal theoretical echo model, and the thickness of the thin film is calculated based on the optimal theoretical waveform.

[0010] Secondly, embodiments of the present invention also provide a film thickness detection device, which is used to execute the film thickness detection method of any embodiment of the present invention, the film thickness detection device comprising:

[0011] The acquisition module is used to acquire data on the reflectivity of probe light on the thin film surface as a function of time, and to generate experimental waveforms.

[0012] The model adjustment module is used to establish a theoretical echo model. The theoretical echo model calculates the theoretical waveforms of the reflectivity of the probe light on the thin film surface as a function of time by setting and adjusting adjustable parameters.

[0013] The model optimization module is used to compare the experimental waveform with multiple theoretical waveforms and select the theoretical echo model that matches the experimental waveform the most, as the optimal theoretical echo model.

[0014] The thickness calculation module is used to calculate the optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time based on the optimal theoretical echo model, and to calculate the thickness of the thin film based on the optimal theoretical waveform.

[0015] This invention provides a method and apparatus for detecting thin film thickness. It acquires data on the reflectivity of probe light on the thin film surface as a function of time, forming an experimental waveform. This experimental waveform is then fitted to a theoretical waveform to obtain an optimal theoretical echo model. By establishing and adjusting adjustable parameters, the theoretical echo model calculates multiple theoretical waveforms showing the reflectivity of probe light on the thin film surface as a function of time. The experimental waveform is compared with these theoretical waveforms, and the theoretical echo model that best matches the experimental waveform is selected as the optimal theoretical echo model. Based on this optimal theoretical echo model, the optimal theoretical waveform showing the reflectivity of probe light on the thin film surface as a function of time is calculated, and the thickness of the thin film is calculated from this optimal theoretical waveform. This method reduces errors in measuring thin film thickness, improves the accuracy and speed of the measurement, ensures the accuracy and reliability of the detection results, and helps improve the yield and efficiency of semiconductor processing. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of an existing photoacoustic film thickness measurement technology.

[0017] Figure 2 This is a schematic diagram of single-layer film echo in the prior art;

[0018] Figure 3 This is a schematic diagram of multilayer film echo technology in the prior art;

[0019] Figure 4A flowchart illustrating a method for detecting thin film thickness provided in an embodiment of the present invention;

[0020] Figure 5 A flowchart of another method for detecting film thickness provided in an embodiment of the present invention;

[0021] Figure 6 A flowchart illustrating a method for calculating thin film thickness provided in an embodiment of the present invention;

[0022] Figure 7 A flowchart illustrating yet another method for detecting film thickness provided in an embodiment of the present invention;

[0023] Figure 8 This is a schematic diagram of the structure of a thin film thickness detection device according to an embodiment of the present invention. Detailed Implementation

[0024] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0025] The core principle of photoacoustic film thickness measurement technology is to use laser pulses to excite the material, causing it to thermally expand and generate sound waves. These sound waves propagate inside the material and are detected. By analyzing the propagation characteristics, propagation time, and sound velocity in the material, the thickness of the film can be deduced.

[0026] Figures 1-3 This is a schematic diagram illustrating the principle of photoacoustic film thickness measurement technology. Please refer to it. Figure 1 The specific process of photoacoustic film thickness measurement technology includes: (1) Laser excitation: using an ultrafast laser pulse to irradiate the surface of the thin film, which may include multiple layers, such as Figure 1Film1 and Film2 in the middle, the thin film is placed on the substrate, so that the material absorbs light energy and converts it into heat energy, causing local thermal expansion and strain of the material; among them, the ultrafast laser pulse can include pump pulse and probe pulse. The pump pulse irradiates the surface of the thin film, causing local thermal expansion and strain of the material; (2) Stress wave generation and propagation: the stress wave generated by the strain will propagate in the thin film and the substrate. The propagation speed and attenuation characteristics of the stress wave are closely related to the thickness of the thin film and the material properties; (3) Reflection and refraction at the interface: when the stress wave propagates to the interface of different thin films, due to the impedance inconsistency, part of the wave is transmitted, and the returned part is called the echo signal; (4) Echo signal detection and analysis: the echo signal is characterized as the change of the reflectivity of the probe light irradiating the surface of the thin film. The echo signal is received by a high-sensitivity detector, and the frequency, amplitude, peak position and propagation time of the signal are analyzed to calculate the thickness of the thin film.

[0027] Figure 2 This is a schematic diagram of a single-layer thin-film echo. Figure 3 This is a schematic diagram of echo signals from a multilayer thin film. In photoacoustic film thickness measurement, the processing and analysis of the echo signal are crucial and key to accurate film thickness measurement. Figure 2 As shown, theoretically, stress waves will repeatedly undergo reflection and transmission effects on the upper and lower surfaces of the monolayer film, thus allowing periodic multiple echo signals to be detected at the surface over a sufficiently long period (e.g., Figure 2 Echo1 and Echo2), this period corresponds to the time it takes for the echo to travel one round trip within this layer. The first echo signal reflected from the lower surface of the monolayer back to the upper surface of the monolayer is also called the first echo signal. For example, Figure 2 Echo1 in the figure represents the first echo signal of film1. Figure 3 Echo11 in the figure represents the first echo signal of film1. Figure 3 Echo21 in the figure represents the first echo signal of the thin film Film2. However, as... Figure 3 As shown, due to crosstalk between multilayer films and attenuation effects from multiple propagations, subsequent effective signals may be suppressed by background noise. Furthermore, multilayer film measurements may encounter overlapping echo signals from different films (e.g., Figure 3 The presence of Echo11 and Echo21 in the Echo21 model presents a challenge for calculating the thickness of multilayer films, making the calculation process quite complex.

[0028] To address these issues, this invention provides a method for detecting thin film thickness. This method is applicable to measuring the thickness of multilayer films using photoacoustic film thickness measurement technology. It derives an optimal theoretical echo model based on the experimental waveform showing the reflectivity of probe light on the thin film surface changing over time, and calculates the film thickness using this optimal theoretical waveform. This reduces measurement errors and improves the accuracy and speed of film thickness measurement. The method can be executed by a thin film thickness detection device, which can be implemented in hardware and / or software. Figure 4 A flowchart of a method for detecting thin film thickness provided in an embodiment of the present invention is shown below. Figure 4 As shown, the methods for detecting film thickness include:

[0029] S110. Obtain data on the reflectivity of the probe light on the thin film surface as a function of time, and form an experimental waveform.

[0030] The thin film can be a semiconductor thin film, and it can also be a non-transparent thin film. The experimental waveform of the reflectivity of the probe light on the thin film surface changing over time can characterize the echo signal within the thin film. The experimental waveform is obtained by processing the original waveform formed by the data of the reflectivity of the probe light on the thin film surface changing over time through unit conversion, background removal, and effective time window truncation.

[0031] Specifically, when measuring the thickness of a thin film using photoacoustic film thickness measurement technology, pump light and probe light need to be irradiated on the film surface. Irradiating the film surface with pump light causes local thermal expansion of the film material and generates strain. The stress wave generated by the strain will propagate in the film and the substrate. The propagation speed and attenuation characteristics of the stress wave are closely related to the thickness and material properties of the film. When the stress wave propagates to the interface between different films, due to the impedance inconsistency, part of the wave is transmitted, and the returned part is called the echo signal. The echo signal is characterized as the change in the reflectivity of the probe light irradiating the film surface. After being received and processed by a high-sensitivity detector, the experimental waveform of the reflectivity of the probe light on the film surface changing with time can be obtained. In order to fit the experimental waveform with the theoretical waveform, the optimal theoretical echo model can be obtained, thereby calculating the film thickness.

[0032] S120. Establish a theoretical echo model. By setting and adjusting adjustable parameters, the theoretical echo model calculates the theoretical waveform of the reflectivity of the probe light on multiple thin film surfaces as a function of time.

[0033] To address issues such as aliasing and signal attenuation that may occur with multilayer films, a theoretical echo model can be established based on the characteristics of the thin film material, the energy of the pump light, and the waveform of the reflectivity of the probe light on the thin film surface as a function of time obtained from multiple experiments. This theoretical echo model can calculate the theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time. For thin films made of different materials, the theoretical waveform calculated by the theoretical echo model may differ from the experimental waveform; therefore, the adjustable parameters in the theoretical echo model need to be adjusted. When measuring the thickness of the same type of thin film, especially when the film is multilayer, existing experimental methods require background removal of overlapping echo signals from different films each time, making the calculation process complex and time-consuming. The theoretical echo model, however, can directly calculate the theoretical waveform of the corresponding film layer, thereby improving the speed of thin film thickness measurement and ensuring the accuracy and reliability of the detection results. Compared to existing measurement methods, calculating thin film thickness using a theoretical echo model helps improve the efficiency of semiconductor processing. Here, "the same type of thin film" refers to thin films with the same material, the same number of layers, but different layer thicknesses.

[0034] Specifically, by repeatedly setting and adjusting the adjustable parameters in the established theoretical echo model, the theoretical waveform calculated by the theoretical echo model can be changed. Therefore, based on the adjusted theoretical echo model, theoretical waveforms showing the reflectivity of probe light on multiple thin film surfaces changing over time can be calculated, with each theoretical waveform corresponding to an adjusted theoretical echo model. By repeatedly adjusting the adjustable parameters in the pre-established theoretical echo model, the fit between the theoretical waveform and the experimental waveform can be maximized, thereby improving the accuracy and reliability of the theoretical waveforms calculated by the theoretical echo model.

[0035] S130. Compare the experimental waveform with multiple theoretical waveforms, and select the theoretical echo model that matches the experimental waveform the most, as the optimal theoretical echo model.

[0036] Among them, Goodness of Fit (GOF) is a statistical method used to evaluate the goodness of fit of a model. It can determine the degree of fit between actual observed data and a theoretical distribution. The higher the GOF, the better the fit between the actual observed data and the theoretical distribution. In this embodiment of the invention, experimental waveforms can be compared with multiple theoretical waveforms, and the theoretical echo model with the highest degree of matching with the experimental waveform can be selected as the optimal theoretical echo model. For example, by calculating the optimal fitness between multiple theoretical waveforms and experimental waveforms, the largest optimal fitness is selected from multiple optimal fitness values, and the adjusted theoretical echo model corresponding to the largest optimal fitness is determined as the optimal theoretical echo model. This can improve the accuracy and reliability of the theoretical waveforms calculated by the optimal theoretical echo model, and provide a reliable basis for the film thickness calculation results.

[0037] Specifically, by calculating the optimal fitness between multiple theoretical and experimental waveforms, the largest optimal fitness is selected from these optimal fitness values. This is achieved by continuously adjusting the adjustable parameters in the pre-established theoretical echo model to maximize the optimal fitness between the theoretical and experimental waveforms. The adjusted theoretical echo model corresponding to the largest optimal fitness is then used as the optimal theoretical echo model. The theoretical waveform calculated using the optimal theoretical echo model has the highest degree of fit with the experimental waveform, ensuring the accuracy and reliability of the theoretical waveform calculated by the optimal theoretical echo model and providing a reliable basis for the film thickness calculation results.

[0038] Furthermore, once the optimal theoretical echo model for the same type of thin film is obtained, when measuring the film thickness of different thin films of the same type, the theoretical waveform can be directly calculated using the optimal theoretical echo model of this invention. This can improve the speed of film thickness measurement and ensure the accuracy and reliability of the detection results. Compared with existing measurement methods, calculating film thickness using the theoretical echo model helps to improve the accuracy and efficiency of semiconductor processing.

[0039] S140. The optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time is calculated based on the optimal theoretical echo model, and the thickness of the thin film is calculated based on the optimal theoretical waveform.

[0040] In existing technologies, when calculating the thickness of a thin film using experimental waveforms, the thickness can usually be calculated using methods such as velocity formulas.

[0041] Specifically, the optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time is calculated based on the optimal theoretical echo model. When calculating the film thickness based on this waveform, methods such as velocity formulas are commonly used. For example, when the film is multilayered, target windows of the echo signals from different layers can be extracted from the optimal theoretical waveform, and the thickness of the corresponding layer can be calculated based on the peak value within the target window and the propagation speed of the sound wave in that layer. When the film is single-layered, the thickness can be directly calculated from the peak value in the optimal theoretical waveform and the propagation speed of the sound wave in the film. When detecting the thickness of the same type of film, calculating the thickness using the optimal theoretical echo model saves on experimental procedures and data processing, significantly improving the speed and efficiency of film thickness measurement, and reducing experimental and labor costs.

[0042] It should be noted that the semiconductor manufacturing industry has stringent requirements for thin film thickness. For example, in integrated circuit manufacturing, precisely controlling the thickness of thin films such as oxide layers and metal layers is crucial to ensuring device performance and stability. The detection method of this invention can be used to measure the thickness of thin films of different materials, such as silicon doped layers, photoresist layers, and dielectric material layers. Photoacoustic film thickness measurement technology, by accurately measuring film thickness, helps semiconductor manufacturers better control the thin film deposition process, ensuring that the thickness of each thin film layer meets design requirements. This directly promotes improved device performance, reduced process defects, and increased yield. Furthermore, the detection method of this invention can measure the thickness of multilayer film structures, improving the detection efficiency of multilayer film structure thickness, which is of great significance for the manufacturing of complex semiconductor devices.

[0043] This invention provides a method for detecting thin film thickness. It acquires data on the reflectivity of probe light on the thin film surface as a function of time, forming an experimental waveform. This experimental waveform is then fitted to a theoretical waveform to obtain an optimal theoretical echo model. By establishing and adjusting adjustable parameters, the theoretical echo model calculates multiple theoretical waveforms showing the reflectivity of probe light on the thin film surface as a function of time. The experimental waveform is compared with these theoretical waveforms, and the theoretical echo model that best matches the experimental waveform is selected as the optimal theoretical echo model. Based on this optimal theoretical echo model, the optimal theoretical waveform showing the reflectivity of probe light on the thin film surface as a function of time is calculated, and the thickness of the thin film is calculated from this optimal theoretical waveform. This method reduces errors in measuring thin film thickness, improves the accuracy and speed of the measurement, ensures the accuracy and reliability of the detection results, and helps improve the yield and efficiency of semiconductor processing.

[0044] This invention also provides another method for detecting film thickness. Figure 5A flowchart of another method for detecting film thickness provided in an embodiment of the present invention is shown below. Figure 5 As shown, the detection methods include:

[0045] S210. Obtain the original waveform of the reflectivity of the probe light on the thin film surface as a function of time.

[0046] Existing photoacoustic film thickness measurement techniques face the challenge of stress waves generated in the thin film repeatedly reflecting and transmitting across the upper and lower surfaces of a single-layer film. This results in the detection of periodic echo signals at the surface over a sufficiently long period, with the period corresponding to the round-trip time of the echo within the layer. However, due to crosstalk between multilayer films and attenuation effects from multiple propagations, subsequent effective signals may be suppressed by background noise. Furthermore, multilayer film measurements may encounter overlapping echo signals from different films, posing a challenge to calculating the multilayer film thickness. Therefore, processing and analysis of the original waveform are necessary for film thickness calculation.

[0047] Specifically, the echo signal in the thin film can be characterized as the original waveform of the reflectivity of the probe light on the thin film surface changing over time. A highly sensitive detector can receive this original waveform. However, the original waveform cannot be directly used for film thickness calculation; it needs to be processed and analyzed to ensure accurate film thickness measurement. For example, there is a time delay between the pump light and the probe light. Focusing the pump light onto the thin film surface causes longitudinal coherent acoustic phonon oscillations, emitting picosecond acoustic pulses into the film. When these picosecond acoustic pulses propagate to the interface between the thin film and air, they generate picosecond acoustic pulse echoes. When the time-delayed probe light reaches the thin film surface, the picosecond acoustic pulse echo modulates the probe light reflected from the thin film surface, forming a picosecond acoustic pulse echo signal, which is the original waveform of the reflectivity of the probe light on the thin film surface changing over time.

[0048] It should be noted that the original waveform contains multiple echo signals, and usually only the first echo signal needs to be obtained from the original waveform for film thickness calculation. When the film contains multiple sub-films, the original waveform contains multiple sub-films corresponding to multiple echo signals. Usually only the first echo signal corresponding to each sub-film in the original waveform needs to be obtained for film thickness calculation. Therefore, further processing of the original waveform is required.

[0049] S220. Perform time zero-point calibration, background subtraction, and time window truncation on the original waveform to obtain the experimental waveform.

[0050] Optionally, the original waveform can be calibrated for zero time, have its background subtracted, and be truncated within a time window to obtain the experimental waveform, specifically including:

[0051] The lowest point of the original waveform before the thermal relaxation process of the thin film is taken as the zero point of time, and the time unit of the horizontal axis in the original waveform is converted. The background of the data after the zero point of time in the original waveform is subtracted.

[0052] The return time of the first echo signal in the film is obtained based on the target thickness of the film and the sound velocity in the film. The region containing the first echo signal in the film in the original waveform after background subtraction is determined as the target window based on the return time. The first echo signal is the signal of the stress wave in the film reflected from the second side of the film and returning to the first side of the film.

[0053] The data of the original waveform after background subtraction in the target window is taken as the experimental waveform.

[0054] Specifically, after acquiring the original waveform data of the reflectivity of the probe light on the thin film surface changing over time, the lowest point of reflectivity change before the thermal relaxation process of the thin film surface is found in the original waveform and taken as the time zero point. The abscissa of the original waveform is converted to picoseconds by combining the movement speed of the delay stage. Next, background subtraction is performed on the data after the time zero point in the original waveform. Generally, a polynomial can be used for fitting; in this embodiment, a double exponential function can be used as the background curve, which can remove the background more cleanly. Simultaneously, the return time of the first echo signal in the thin film can be estimated based on the target thickness of the thin film and the sound velocity in the thin film. Based on the return time, the region containing the first echo signal in the thin film in the original waveform after background subtraction is determined as the target window. The data of the original waveform after background subtraction within the target window is extracted as the experimental waveform. Furthermore, when the thin film is a multilayer film, the return time required for the first echo signal in each film layer can be estimated by combining the number of film layers, the target thickness of each film layer, and the propagation speed of different materials in different layers of the film. Based on the return time required for the first echo signal in each film layer, the region of the first echo signal of the corresponding film layer in the original waveform after background subtraction is determined as the target window corresponding to different film layers. By extracting the data of the target window corresponding to different film layers in the original waveform after background subtraction, the experimental waveform corresponding to different film layers is obtained, thereby realizing the processing of experimental data of multilayer thin films.

[0055] In existing technologies, after selecting an effective time window, the maximum or minimum value of the target window (determined by the sign of the reflection coefficient) can be directly found by solving the curve slope as the first echo signal. After obtaining the echo time t based on the corresponding abscissa, the displacement-velocity formula is used... Calculating film thickness is a complex process, but the drawback of existing technologies is that when measuring the thickness of the same type of film, each measurement requires background removal operations such as time zero-point calibration, background subtraction, and time window truncation of the original waveform. This results in low efficiency in film thickness detection.

[0056] S230. Establish a theoretical echo model.

[0057] In existing photoacoustic film thickness measurement technologies, when measuring the thickness of the same type of thin film, background removal operations such as time zero-point calibration, background subtraction, and time window truncation are required for each measurement of the original waveform, resulting in low measurement efficiency and negatively impacting the yield and efficiency of semiconductor processing. This invention addresses this by establishing a theoretical echo model for theoretical waveform calculation and automating the algorithmic processing of experimental data, thereby improving the accuracy and speed of thin film measurement and contributing to enhanced yield and efficiency in semiconductor processing.

[0058] Specifically, in establishing the theoretical echo model, a double-exponential function is used as the background curve, which can eliminate the background more cleanly. The specific process includes: establishing the sensitivity function f(z) and establishing the strain function η. 33 (z, t), where, firstly, the sensitivity function f(z) is defined to satisfy:

[0059]

[0060] in, n represents the real part of the refractive index of the thin film material, and κ represents the imaginary part of the refractive index of the thin film material. and All parameters are adjustable. Characterizing the degree to which the real part of the refractive index of a thin film material changes with strain. The imaginary part of the refractive index of a thin film material varies with strain, where λ represents the wavelength of the probe light, ω is the angular frequency of the probe light, c is the speed of light, and z represents the longitudinal depth from the bottom to the surface of the film. Due to adjustable parameters... and None of them can be calculated or measured by theory or experiment, so they are adjusted as adjustable parameters in the subsequent fitting process.

[0061] Then, define the strain function η. 33 (z, t) satisfies:

[0062]

[0063] Where R is the reflectivity of light irradiating the thin film surface, Q is the pulse energy of light, β is the linear expansion coefficient of the thin film material, A is the size of the light spot formed when light irradiates the thin film surface. It should be noted that the light here can specifically be the light that causes strain in the thin film, such as pump light, δ is the penetration depth of the thin film material, C is the specific heat capacity of the thin film, and υ is the Poisson's ratio of the thin film material.

[0064] Finally, the sensitivity function f(z) and the strain function η 33 The theoretical echo model is obtained by convolving (z, t), and the theoretical echo model is represented by the convolution function ΔR.

[0065] Specifically, the sensitivity function f(z) and the strain function η can be used to... 33 The convolution function ΔR obtained after convolving (z, t) satisfies:

[0066]

[0067] The convolution function ΔR is used as the theoretical echo model. This model allows for the calculation of the theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time, thereby increasing the speed of thin film thickness measurement and ensuring the accuracy and reliability of the detection results. Compared to existing measurement methods, calculating thin film thickness using the theoretical echo model helps improve the efficiency of semiconductor processing.

[0068] S240. Adjust the adjustable parameters in the theoretical echo model multiple times, and calculate the theoretical waveform of the reflectivity of the probe light on multiple thin film surfaces as a function of time based on the adjusted theoretical echo model.

[0069] Specifically, adjustable parameters and None of these can be calculated or measured using theory or experiments; it is necessary to continuously adjust the adjustable parameters in the theoretical echo model. and Adjustments are made to alter the theoretical waveform calculated by the theoretical echo model. This involves calculating the fitness between multiple adjusted theoretical and experimental waveforms to maximize the fitness between them, thereby improving the accuracy and reliability of the theoretical waveform calculated by the theoretical echo model. Optional adjustable parameters include... and The range of variation is [-1, 1].

[0070] S250. Calculate the optimal fitness among multiple theoretical waveforms and experimental waveforms, select the largest optimal fitness from multiple optimal fitness, and determine the adjusted theoretical echo model corresponding to the largest optimal fitness as the optimal theoretical echo model.

[0071] Optional, adjust adjustable parameters and Multiple theoretical waveforms were obtained, and the theoretical waveforms were compared with the experimental waveforms to obtain the adjustable parameters. and Within a reasonable range.

[0072] Within a reasonable range, continue adjusting the adjustable parameters and calculate the optimal fitness (GOF) between multiple theoretical and experimental waveforms. The optimal fitness GOF satisfies:

[0073]

[0074] Where N represents the number of time sampling points within the target window, deltaR exp DeltaR represents the change in reflectance in the experimental waveform. cal DeltaR represents the change in reflectivity in the theoretical waveform. cal_mean This represents the average value of the reflectance variation in the original waveform within the target window; an adjustable parameter. and The reasonable range is [-1, 1].

[0075] The theoretical echo model corresponding to the maximum value of the optimal fitness GOF is taken as the optimal theoretical echo model.

[0076] Among them, the optimal fitness (GOF) characterizes the degree of matching between theory and experiment; the larger the GOF and the closer it is to 1, the higher the degree of matching. Different thin film materials correspond to different optimal adjustable parameters.

[0077] Specifically, by continuously comparing experimental waveforms with theoretical waveforms, reasonable ranges for various parameters of different thin film materials are obtained, and then the adjustable parameters are gradually adjusted within these reasonable ranges. and The optimal fitness of light (GOF) between multiple theoretical and experimental waveforms is calculated. The highest GOF is selected, and the adjustable parameter in the adjusted theoretical echo model corresponding to the highest GOF is determined as the optimal adjustable parameter. Different thin film materials correspond to different optimal adjustable parameters; the theoretical echo model corresponding to the optimal adjustable parameter is taken as the optimal theoretical echo model. The theoretical waveform calculated using the optimal theoretical echo model has the highest degree of fit with the experimental waveform, ensuring the accuracy and reliability of the theoretical waveform calculated by the optimal theoretical echo model, providing a reliable basis for thin film thickness calculations. When calculating the thickness of multiple thin films of the same type using the optimal theoretical echo model, photoacoustic film thickness measurement experiments are unnecessary. Only the material properties of the thin film and other fixed experimental data are needed to calculate the theoretical waveforms of different thin films, and then the film thickness can be calculated based on the theoretical waveforms. Compared with existing technologies, this significantly improves the accuracy and speed of thin film thickness measurement, while ensuring the accuracy and reliability of the detection results, contributing to improved yield and efficiency in semiconductor processing.

[0078] The optimal fitness is selected from multiple optimal fitness values, and the adjusted theoretical echo model corresponding to the maximum optimal fitness is determined as the optimal theoretical echo model.

[0079] S260. The optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time is calculated based on the optimal theoretical echo model, and the thickness of the thin film is calculated based on the optimal theoretical waveform.

[0080] Among them, the theoretical waveform calculated by the optimal theoretical echo model has the highest degree of fit with the experimental waveform. The thickness of the thin film calculated by the theoretical waveform calculated by the optimal theoretical echo model can be regarded as the actual thickness of the thin film. Therefore, the theoretical waveform of the reflectivity of the probe light on the thin film surface changing with time calculated by the optimal theoretical echo model is taken as the optimal theoretical waveform, and the thickness of the thin film calculated by the optimal theoretical waveform is the actual thickness of the thin film.

[0081] This invention also provides a method for calculating film thickness. Figure 6 A flowchart illustrating a method for calculating thin film thickness provided in an embodiment of the present invention is shown below. Figure 6 As shown, it specifically includes:

[0082] S310. Take the maximum or minimum value in the optimal theoretical waveform as the first echo signal, and determine the echo time corresponding to the first echo signal in the optimal theoretical waveform; wherein, each sub-membrane corresponds to an optimal theoretical waveform, and each optimal theoretical waveform contains the first echo signal in the corresponding sub-membrane.

[0083] Specifically, when the thin film comprises multiple sub-films, each sub-film corresponds to an optimal theoretical waveform. Each optimal theoretical waveform contains the first echo signal of the corresponding sub-film. For example, when the thin film has only one layer, the thin film has only one corresponding optimal theoretical echo model. The optimal theoretical waveform showing the time-varying reflectivity of the probe light on the thin film surface can be calculated using this optimal theoretical echo model. This optimal theoretical waveform contains the first echo signal of the thin film. When the thin film has multiple layers, including multiple sub-films, each sub-film corresponds to an optimal theoretical echo model. The optimal theoretical waveform corresponding to the corresponding sub-film can be calculated using the optimal theoretical waveforms of different sub-films. This optimal theoretical waveform contains the first echo signal of the corresponding sub-film. The echo signal of the sub-film is the signal of the stress wave returning from the bottom of the sub-film to the surface of the uppermost thin film layer. By using the maximum or minimum value in the optimal theoretical waveform as the first echo signal and determining the echo time corresponding to the first echo signal in the optimal theoretical waveform, the thickness of the corresponding sub-film can be calculated according to the velocity formula.

[0084] S320. The thickness of the sub-membrane corresponding to the target window in the optimal theoretical waveform is calculated based on the echo time and the acoustic wave propagation speed in the corresponding sub-membrane.

[0085] Specifically, based on the velocity formula d = vt / 2, the thickness of the sub-film corresponding to the target window in the optimal theoretical waveform is calculated.

[0086] Where d is the thickness of the sub-membrane corresponding to the optimal theoretical waveform, v is the sound wave propagation speed in the sub-membrane, and t is the echo time.

[0087] The thickness of the thin film calculated by the thin film thickness detection method in this embodiment of the invention can be considered as the actual thickness of the thin film. This method eliminates the need for multiple experiments to obtain data, reduces the error in measuring the thin film thickness, improves the accuracy and speed of the measurement, ensures the accuracy and reliability of the detection results, and helps to improve the yield and efficiency of the semiconductor processing flow.

[0088] This invention also provides another method for detecting film thickness, illustrating the design flow of such a method. Figure 7 A flowchart of another method for detecting film thickness provided in an embodiment of the present invention is shown below. Figure 7 As shown, it specifically includes:

[0089] S410, Experimental Data Processing.

[0090] Specifically, the experimental data processing can include converting the time coordinate unit, redefining the time zero point, subtracting the background by fitting the function, finding the peak position of the multilayer film, and measuring the film thickness of the original waveform of the reflectivity of the probe light on the thin film surface as a function of time. The experimental waveform of the reflectivity of the probe light on the thin film surface as a function of time can be obtained after the experimental data processing steps. By fitting the experimental waveform and the theoretical waveform, the optimal theoretical echo model is obtained.

[0091] S420, Theoretical echo model established.

[0092] Specifically, the theoretical echo model can combine the sensitivity function f(z) and the strain function η. 33 The result is obtained after convolution of (z, t).

[0093] S430, Adjustment of intrinsic parameters of theoretical echo model.

[0094] Specifically, the adjustable parameters in the pre-established theoretical echo model were adjusted multiple times, and the theoretical waveforms of the reflectivity of the probe light on multiple thin film surfaces as a function of time were calculated based on the adjusted theoretical echo model.

[0095] S440. Determine the optimal theoretical echo model.

[0096] Specifically, by calculating the optimal fitness between multiple theoretical and experimental waveforms, the goodness of fit between the experiment and the model can be given, providing a reliable basis for the calculation results. The optimal fitness is selected from multiple optimal fitness values, and the adjusted theoretical echo model corresponding to the maximum optimal fitness is taken as the optimal theoretical echo model.

[0097] S450, Calculate the film thickness.

[0098] Specifically, this invention addresses issues such as echo signal aliasing and signal attenuation that may occur in multilayer films by providing an optimal theoretical echo model. This model allows for the calculation of the actual film thickness of each layer. Based on the optimal theoretical echo model, the optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time can be calculated, and the film thickness can be calculated from this optimal waveform. Furthermore, automated algorithms are used to process experimental data, thereby improving measurement accuracy and speed, and enhancing the yield and efficiency of semiconductor processing.

[0099] This invention provides a method for detecting thin film thickness. Using an optimal theoretical echo model, the method calculates the optimal theoretical waveform of the reflectivity of the probe light on the thin film surface over time, and then calculates the film thickness based on this optimal theoretical waveform. This reduces measurement errors, improves the accuracy and speed of thin film thickness measurement, and ensures the accuracy and reliability of the detection results, thus contributing to improved yield and efficiency in semiconductor manufacturing processes. This invention helps semiconductor manufacturers better control thin film deposition processes, ensuring that the thickness of each thin film layer meets design requirements. This directly promotes improved device performance, reduced process defects, and increased yield. Furthermore, the detection method of this invention can measure the thickness of multilayer film structures, improving the detection efficiency of multilayer film structures, which is of great significance for the manufacture of complex semiconductor devices.

[0100] This invention also provides a device for detecting film thickness. Figure 8 This is a schematic diagram of a thin film thickness detection device according to an embodiment of the present invention. The thin film thickness detection device in this embodiment is used to execute the thin film thickness detection method in any embodiment of the present invention, such as... Figure 8 As shown, the thin film thickness detection device includes:

[0101] The acquisition module 110 is used to acquire data on the reflectivity of the probe light on the thin film surface as a function of time, and to form an experimental waveform.

[0102] The model adjustment module 120 is used to establish a theoretical echo model. The theoretical echo model calculates the theoretical waveform of the reflectivity of the probe light on multiple thin film surfaces as a function of time by setting and adjusting adjustable parameters.

[0103] The model optimization module 130 is used to compare the experimental waveform with multiple theoretical waveforms and select the theoretical echo model that matches the experimental waveform the most, as the optimal theoretical echo model.

[0104] The thickness calculation module 140 is used to calculate the optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time based on the optimal theoretical echo model, and to calculate the thickness of the thin film based on the optimal theoretical waveform.

[0105] Specifically, the acquisition module 110 receives and processes the experimental waveform of the reflectivity of the probe light on the thin film surface changing over time using a high-sensitivity detector. This waveform is then fitted to the theoretical waveform to obtain the optimal theoretical echo model, thereby calculating the thin film thickness. The model adjustment module 120, through the established theoretical echo model and its adjustable parameters, can change the theoretical waveform calculated by the theoretical echo model. Therefore, based on the adjusted theoretical echo model, multiple theoretical waveforms of the reflectivity of the probe light on the thin film surface changing over time can be calculated, with each theoretical waveform corresponding to an adjusted theoretical echo model. The model optimization module 130 compares the experimental waveform with multiple theoretical waveforms and selects the theoretical echo model that best matches the experimental waveform as the optimal theoretical echo model. Specifically, it calculates the optimal fitness between multiple theoretical waveforms and experimental waveforms, selects the largest optimal fitness from among them, that is, by continuously adjusting the adjustable parameters in the pre-established theoretical echo model to maximize the optimal fitness between the theoretical waveform and the experimental waveform, and then selects the adjusted theoretical echo model corresponding to the largest optimal fitness as the optimal theoretical echo model. The thickness calculation module 140 calculates the optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time based on the optimal theoretical echo model, and calculates the thickness of the thin film based on the optimal theoretical waveform. This saves on the experimental process and data processing, and greatly improves the speed and efficiency of measuring the thickness of the thin film.

[0106] Furthermore, the thin film thickness detection device of the present invention includes, but is not limited to, the above modules. The thin film thickness detection device of the present invention can be adaptively configured with relevant functional modules according to the thin film thickness detection method of any embodiment of the present invention to realize the function and technical effect of the thin film thickness detection method of any embodiment of the present invention.

[0107] This invention provides a thin film thickness detection device. The device acquires data on the reflectivity of probe light on the thin film surface as a function of time, forming an experimental waveform. This experimental waveform is then fitted to a theoretical waveform to obtain an optimal theoretical echo model. A model adjustment module establishes the theoretical echo model, which calculates multiple theoretical waveforms of the reflectivity of probe light on the thin film surface as a function of time by setting and adjusting adjustable parameters. A model optimization module compares the experimental waveform with these theoretical waveforms, selecting the theoretical echo model with the highest degree of matching as the optimal theoretical echo model. A thickness calculation module calculates the optimal theoretical waveform of the reflectivity of probe light on the thin film surface as a function of time based on the optimal theoretical echo model, and calculates the thickness of the thin film based on the optimal theoretical waveform. This reduces the error in measuring thin film thickness, improves the accuracy and speed of the measurement, ensures the accuracy and reliability of the detection results, and helps improve the yield and efficiency of semiconductor processing.

[0108] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for detecting film thickness, characterized in that, The method includes: Data on the reflectivity of probe light on the thin film surface as a function of time were obtained to form experimental waveforms; A theoretical echo model is established, which calculates the theoretical waveforms of the reflectivity of the probe light on the thin film surface as a function of time by setting and adjusting adjustable parameters. The experimental waveform is compared with multiple theoretical waveforms, and the theoretical echo model that matches the experimental waveform the most is selected as the optimal theoretical echo model. The optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time is calculated based on the optimal theoretical echo model, and the thickness of the thin film is calculated based on the optimal theoretical waveform. The establishment of the theoretical echo model includes establishing a sensitivity function. and establishing strain function ,in: Define the sensitivity function : ； in, , n and κ are the real and imaginary parts of the refractive index of the thin film material, respectively. The adjustable parameter, Characterizing the degree to which the real part of the refractive index of the thin film material changes with strain. The imaginary part of the refractive index of the thin film material changes with strain, where λ represents the wavelength of the probe light. Here, c is the angular frequency of the probe light, c is the speed of light, and z represents the longitudinal depth from the bottom of the film to the surface of the film. Define the strain function : ； Where R is the reflectivity of light irradiating the surface of the thin film, Q is the pulse energy of light, β is the coefficient of linear expansion of the thin film material, A is the size of the light spot formed when light irradiates the surface of the thin film, δ is the penetration depth of the thin film material, C is the specific heat capacity of the thin film, and υ is the Poisson's ratio of the thin film material; The sensitivity function and the strain function The theoretical echo model is obtained after convolution; The theoretical echo model uses a convolution function. This indicates that the convolution function is defined. : 。 2. The method for detecting film thickness according to claim 1, characterized in that, The process of acquiring data on the reflectivity of the probe light on the thin film surface as a function of time, forming experimental waveforms, includes: Obtain the original waveform of the reflectivity of the probe light on the surface of the thin film as a function of time; The original waveform is subjected to time zero-point calibration, background subtraction, and time window truncation to obtain the experimental waveform.

3. The method for detecting film thickness according to claim 2, characterized in that, The experimental waveform is obtained by performing time zero-point calibration, background subtraction, and time window truncation on the original waveform, including: The lowest point of the original waveform before the thermal relaxation process of the thin film is taken as the time zero point, and the time unit of the horizontal axis in the original waveform is converted. The data after the time zero point in the original waveform is deducted from the background. The return time of the first echo signal in the film is obtained based on the target thickness of the film and the sound velocity in the film. The region containing the first echo signal in the film in the original waveform after background subtraction is determined as the target window based on the return time. The first echo signal is the signal of the stress wave in the film reflected from the second side of the film and returning to the first side of the film. The data of the original waveform after background subtraction is extracted in the target window and used as the experimental waveform.

4. The method for detecting film thickness according to claim 3, characterized in that, The optimal theoretical echo models selected include: Adjust the adjustable parameters Multiple theoretical waveforms are obtained, and the multiple theoretical waveforms are compared with the experimental waveforms to obtain the adjustable parameters. Within a reasonable range; Within the reasonable range, the adjustable parameters are further adjusted to calculate the optimal fitness between multiple theoretical waveforms and experimental waveforms. The optimal fitness satisfy: ； Where N represents the number of time sampling points within the target window. This represents the change in reflectivity in the experimental waveform. This represents the change in reflectivity in the theoretical waveform. This represents the average value of the reflectance change in the original waveform within the target window; The optimal fitness The theoretical echo model corresponding to the maximum value is taken as the optimal theoretical echo model.

5. The method for detecting film thickness according to claim 4, characterized in that, The adjustable parameters The reasonable range mentioned above is all .

6. The method for detecting film thickness according to claim 1, characterized in that, The thin film comprises multiple sub-films; the thickness of the thin film is calculated based on the optimal theoretical waveform, including: The maximum or minimum value in the optimal theoretical waveform is taken as the first echo signal, and the echo time corresponding to the first echo signal in the optimal theoretical waveform is determined; wherein, each of the sub-films corresponds to one optimal theoretical waveform, and each optimal theoretical waveform contains the first echo signal in the corresponding sub-film; The thickness of the sub-membrane corresponding to the optimal theoretical waveform is calculated based on the echo time and the acoustic wave propagation speed in the corresponding sub-membrane.

7. The method for detecting film thickness according to claim 6, characterized in that, The thickness of the sub-membrane corresponding to the target window in the optimal theoretical waveform is calculated based on the echo time and the acoustic wave propagation velocity in the corresponding sub-membrane, including: Based on the velocity formula d=vt / 2, the thickness of the sub-film corresponding to the target window in the optimal theoretical waveform is calculated. Where d is the thickness of the sub-membrane corresponding to the optimal theoretical waveform, v is the sound wave propagation speed in the sub-membrane, and t is the echo time.

8. A device for detecting thin film thickness, characterized in that, The thin film thickness detection device includes the thin film thickness detection method according to any one of claims 1-7, wherein the thin film thickness detection device includes: The acquisition module is used to acquire data on the reflectivity of the probe light on the thin film surface as a function of time, and to generate experimental waveforms. The model adjustment module is used to establish a theoretical echo model. The theoretical echo model calculates the theoretical waveforms of the reflectivity of the probe light on the thin film surface as a function of time by setting and adjusting adjustable parameters. The model optimization module is used to compare the experimental waveform with multiple theoretical waveforms and select the theoretical echo model that matches the experimental waveform the most, as the optimal theoretical echo model. The thickness calculation module is used to calculate the optimal theoretical waveform of the reflectivity of the probe light on the thin film surface as a function of time based on the optimal theoretical echo model, and to calculate the thickness of the thin film based on the optimal theoretical waveform.

Images