Method for establishing spectrum library, method and device for monitoring thickness and polishing equipment

Abstract

The invention relates to a method for establishing a spectrum library, a method and device for monitoring thickness and polishing equipment. The method for establishing the spectrum library comprises the following steps: acquiring an actually measured energy reflectivity spectrum of a sample wafer based on an optical detection module, acquiring a first electric field reflectivity of incident light of the optical detection module on an interface between the grinding liquid and a first grinding layer of the target sample wafer and a second electric field reflectivity of the incident light on an interface between the first grinding layer and a first substrate layer of the target sample wafer; based on the first electric field reflectivity, the second electric field reflectivity, the wavelength of the incident light, the refractive index of the incident light on the grinding layer and the thickness of the grinding layer, establishing a theoretical total electric field reflectivity model; obtaining a theoretical energy reflectivity model based on the theoretical total electric field reflectivity model; based on the theoretical energy reflectivity model, theoretical energy reflectivity spectrums corresponding to different grinding layer thicknesses are calculated to generate a spectrum library. According to the invention, the efficiency of establishing the spectrum library can be improved, and meanwhile, a finer spectrum library can be obtained.

Description

Technical Field

[0001] This disclosure relates to the field of semiconductor technology, and in particular to a method for establishing a spectral library, a method for monitoring thickness, an apparatus, and a polishing device. Background Technology

[0002] In the manufacturing process of integrated circuit devices, chemical mechanical polishing (CMP) combines chemical etching and mechanical abrasion to remove excess material or reduce film layers to a specified thickness. For metal thin films (such as copper), CMP is typically used to remove excess material; for dielectric thin films (such as silicon oxide and silicon nitride), CMP is typically used to planarize the film surface.

[0003] End Point Detection (EPD) technology based on spectral analysis can be used to monitor the thickness of the polished layer during CMP (Continuous Metallurgy). This technology is based on the principle of optical interference. It uses a broadband light source covering the visible light spectrum as the measurement light to illuminate the thin film surface. The reflected light from the upper interface of the thin film interferes with the reflected light from the lower interface to form a reflection spectrum. A detector with spectral detection capabilities is then used to obtain the reflection spectrum information related to the film thickness.

[0004] In related technologies, before grinding product samples, standard samples need to be ground, and spectra corresponding to different grinding layer thicknesses need to be collected to form a spectral library. Therefore, during the grinding process of the product sample, the corresponding thickness is matched from the spectral library based on the collected spectrum of the product sample, and this is taken as the current grinding layer thickness of the product sample. Since the wear level of the grinding pads in the grinding equipment varies daily, the spectral library needs to be updated daily or every other day to match the real-time working status of the CMP equipment. Furthermore, standard samples need to be ground in advance before each update, resulting in low efficiency. Summary of the Invention

[0005] To overcome the problems existing in related technologies, this disclosure provides a method for establishing a spectral library, a method for monitoring thickness, an apparatus, and a polishing device.

[0006] This disclosure provides a method for establishing a spectral library including several theoretical energy reflectance spectra. The spectral library is used for film thickness monitoring during a chemical mechanical polishing (CMP) process on a target sample. The film thickness monitoring includes acquiring the measured energy reflectance spectrum of the target sample using an optical detection module. The method is characterized by: based on the measured energy reflectance spectrum of the sample acquired by the optical detection module, obtaining the first electric field reflectance of the incident light at the interface between the polishing slurry and the first polishing layer of the target sample, and the reflectance of the incident light at the interface between the first polishing layer and the target sample. The second electric field reflectivity of the interface between the first substrate layers; based on the first electric field reflectivity, the second electric field reflectivity, the wavelength of the incident light, the refractive index of the incident light in the polishing layer, and the thickness of the polishing layer, a theoretical total electric field reflectivity model is established after the incident light is reflected by the target sample; based on the theoretical total electric field reflectivity model, a theoretical energy reflectivity model is obtained after the incident light is reflected by the target sample; based on the theoretical energy reflectivity model, theoretical energy reflectivity spectra corresponding to different polishing layer thicknesses are calculated for use in film thickness monitoring, so as to generate the spectral library.

[0007] In some embodiments, obtaining the first electric field reflectance and the second electric field reflectance based on the measured reflectance spectrum of the sample acquired by the optical detection module includes: preparing a calibration sample, the calibration sample comprising a second polishing layer having a first complex refractive index and a second substrate layer having a second complex refractive index, the second polishing layer being made of the same material as the first polishing layer of the target sample; introducing the polishing slurry into the surface of the calibration sample, and acquiring the corresponding first measured energy reflectance spectrum at at least three measurement locations on the calibration sample based on the optical detection module; determining the third electric field reflectance of the incident light at the interface between the second polishing layer and the second substrate layer based on the first complex refractive index and the second complex refractive index; and measuring the third electric field reflectance of the incident light at the interface between the second polishing layer and the second substrate layer on the calibration sample. The thickness of the second polishing layer at the specified location is measured; based on the third electric field reflectivity, the wavelength of the incident light, the first measured energy reflectivity spectrum, and the corresponding thickness of the second polishing layer, the first electric field reflectivity is determined by simultaneous equations or regression fitting; the polishing slurry is introduced into the surface of the target sample, and the corresponding second measured energy reflectivity spectra at at least three preset locations within the same characteristic region on the target sample are collected based on the optical detection module; the thickness of the first polishing layer at the preset location on the target sample is measured; based on the first electric field reflectivity, the wavelength of the incident light, the second measured energy reflectivity spectrum, and the corresponding thickness of the first polishing layer, the second electric field reflectivity is determined by simultaneous equations or regression fitting.

[0008] In some embodiments, obtaining the first electric field reflectance and the second electric field reflectance based on the measured reflectance spectrum of the sample acquired by the optical detection module includes: preparing a calibration sample, the calibration sample comprising a second polishing layer having a first complex refractive index and a second substrate layer having a second complex refractive index, the second polishing layer being made of the same material as the first polishing layer of the target sample; introducing the polishing slurry into the surface of the calibration sample, and acquiring the corresponding first measured energy reflectance spectrum at at least three measurement locations on the calibration sample based on the optical detection module; determining the third electric field reflectance of the incident light at the interface between the second polishing layer and the second substrate layer based on the first complex refractive index and the second complex refractive index; measuring the corresponding thickness of the second polishing layer at the measurement locations on the calibration sample; and based on the third... The electric field reflectivity, the wavelength of the incident light, the first measured energy reflectivity spectrum, and the corresponding thickness of the second polishing layer are used to determine the first electric field reflectivity by solving simultaneous equations or regression fitting. Without introducing the polishing slurry, the optical detection module collects the corresponding third measured energy reflectivity spectra at at least three preset locations within the same characteristic region on the target sample. Based on the complex refractive index of air and the first complex refractive index, the fourth electric field reflectivity at the interface between the incident light and the first polishing layer is determined. The corresponding thickness of the first polishing layer at the preset location on the target sample is measured. Based on the fourth electric field reflectivity, the wavelength of the incident light, the third measured energy reflectivity spectrum, and the corresponding thickness of the first polishing layer, the second electric field reflectivity is determined by solving simultaneous equations or regression fitting.

[0009] In some embodiments, obtaining the first electric field reflectance and the second electric field reflectance based on the measured reflectance spectrum of the sample acquired by the optical detection module includes: acquiring the corresponding third measured energy reflectance spectrum at at least three preset positions within the same characteristic region on the target sample based on the optical detection module without introducing the polishing slurry; determining the fourth electric field reflectance of the incident light at the interface between the air and the first polishing layer based on the complex refractive index of air and the first complex refractive index of the first polishing layer; measuring the corresponding thickness of the first polishing layer at the preset positions on the target sample; based on The fourth electric field reflectivity, the wavelength of the incident light, the third measured energy reflectivity spectrum, and the corresponding thickness of the first polishing layer are used to determine the second electric field reflectivity by solving simultaneous equations or regression fitting; the polishing slurry is introduced into the surface of the target sample, and the corresponding second measured energy reflectivity spectra at at least three preset positions within the same characteristic region on the target sample are collected based on the optical detection module; the first electric field reflectivity is determined by solving simultaneous equations or regression fitting based on the second electric field reflectivity, the wavelength of the incident light, the second measured energy reflectivity spectrum, and the corresponding thickness of the first polishing layer.

[0010] In some embodiments, before calculating the theoretical energy reflectance spectra corresponding to different polishing layer thicknesses for use in the film thickness monitoring, the method further includes setting multiple polishing layer thicknesses within a preset thickness range with a preset thickness step.

[0011] In some embodiments, the target sample has multiple feature regions, each feature region corresponding to a first electric field reflectance and a second electric field reflectance, and the spectral library includes several theoretical reflectance spectra corresponding to each feature region.

[0012] This disclosure provides a method for monitoring the thickness of a polishing layer during chemical mechanical polishing, comprising: acquiring the current measured energy reflectance spectrum of a target sample in real time based on an optical detection module during the polishing process, wherein the target sample includes a polishing layer and a substrate layer; determining a target theoretical energy reflectance spectrum that matches the current measured energy reflectance spectrum from a spectral library; and determining the thickness value corresponding to the target theoretical energy reflectance spectrum as the current thickness of the polishing layer; wherein the spectral library is established according to any of the above methods.

[0013] In some embodiments, the current measured energy reflectance spectrum is the energy reflectance spectrum after background noise removal processing of the initial energy reflectance spectrum value collected based on the optical detection module.

[0014] In some embodiments, determining the target theoretical energy reflectance spectrum that matches the current measured energy reflectance spectrum from the spectral library includes: determining the current number of rotations of the bearing head used to grind the target sample, and the target feature region corresponding to the current spectral acquisition position; determining the target grinding layer thickness range corresponding to the current number of rotations based on the correspondence between the number of rotations of the bearing head and the grinding layer thickness range; and determining the target theoretical energy reflectance spectrum that matches the current measured energy reflectance spectrum from the theoretical energy reflectance spectra corresponding to the target feature region and the target grinding layer thickness range in the spectral library.

[0015] This disclosure provides an apparatus for monitoring the thickness of a polishing layer on a sample during chemical mechanical polishing, comprising: an optical detection module for real-time acquisition of the current measured energy reflectance spectrum of a target sample during the polishing process, wherein the target sample includes the polishing layer and a substrate layer; and a calculation and processing module for establishing a spectral library and determining a target theoretical energy reflectance spectrum that matches the current measured energy reflectance spectrum from the spectral library, and determining the thickness value corresponding to the target theoretical energy reflectance spectrum as the current thickness of the polishing layer; wherein the spectral library is determined according to any of the methods described above.

[0016] This disclosure provides a chemical mechanical polishing apparatus, including the aforementioned device for monitoring the thickness of the polished layer on a sample.

[0017] The technical solutions provided by the embodiments of this disclosure can include the following beneficial effects: Based on the measured energy reflectance spectrum of the sample acquired by the optical detection module, the first electric field reflectance of the incident light at the interface between the polishing slurry and the first polishing layer of the target sample and the second electric field reflectance of the incident light at the interface between the first polishing layer and the first substrate layer of the target sample are obtained. The theoretical energy reflectance corresponding to different polishing layer thicknesses can be determined by the first and second electric field reflectances to generate a spectral library. Since the first and second electric field reflectances can be accurately calculated by back-deriving from the measured energy reflectance spectra of different samples and under different test environments, the theoretical energy reflectance corresponding to different polishing layer thicknesses can be calculated based on the accurate values ​​of the first and second electric field reflectances, thereby generating a spectral library with flexible and adjustable data, improving the efficiency of establishing the spectral library. That is, the present invention can establish a theoretical energy reflectance spectral library corresponding to arbitrary thickness step size and thickness range, obtaining a more refined spectral library, while solving the problem of difficulty in solving theoretical modeling.

[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0020] Figure 1 This is a schematic diagram of a CMP device containing an EPD system, as shown in an example.

[0021] Figure 2 This is a flowchart illustrating a method for establishing a spectral library comprising several theoretical energy reflectance spectra according to an exemplary embodiment.

[0022] Figure 3 This is a schematic diagram illustrating the first and second electric field reflectivities according to an example.

[0023] Figure 4 This is a flowchart illustrating, according to an exemplary embodiment, a method for acquiring a first electric field reflectance and a second electric field reflectance by acquiring a measured reflectance spectrum of a sample using an optical detection module.

[0024] Figure 5 This is a flowchart illustrating, according to an exemplary embodiment, a method for acquiring a first electric field reflectance and a second electric field reflectance by acquiring a measured reflectance spectrum of a sample using an optical detection module.

[0025] Figure 6 This is a flowchart illustrating, according to an exemplary embodiment, a method for acquiring a first electric field reflectance and a second electric field reflectance by acquiring a measured reflectance spectrum of a sample using an optical detection module.

[0026] Figure 7 This is a flowchart illustrating a method for monitoring the thickness of an abrasive layer during chemical mechanical polishing, according to an exemplary embodiment.

[0027] Figure 8 This is a block diagram of an apparatus for monitoring the thickness of a polished layer on a sample during a chemical mechanical polishing process, according to an exemplary embodiment. Detailed Implementation

[0028] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure.

[0029] With technological advancements, the requirements for planarization in the manufacturing process of integrated circuit devices are becoming increasingly stringent, while the patterns on the substrates are also becoming more complex. To achieve more reliable device performance, it is necessary to monitor the film thickness during the CMP process in order to appropriately control and stop the removal process.

[0030] Among related technologies, End Point Detection (EPD) based on spectral analysis is currently the mainstream method for monitoring the thickness of the polishing layer in the CMP process. It mainly infers the thickness of the polishing layer during the polishing process based on the reflectance spectrum.

[0031] refer to Figure 1 The schematic diagram shown is of a CMP device containing an EPD system. The CMP device containing an EPD system includes: a rotatable grinding table and a grinding pad located on the grinding table, wherein the grinding pad provides an optical path with high transmittance through a window; a grinding slurry spraying arm located on the grinding table for spraying grinding slurry during the grinding process; and an EPD optical detection system located in the internal cavity of the grinding table.

[0032] During the polishing process, a polishing slurry of a certain thickness exists between the sample under test and the polishing pad. When the sample under test passes through the window, the EPD system detects the spectral information of its surface film. Specifically, the incident light from the optical fiber first passes through the window, then through the polishing slurry, and then shines on the sample under test. After being reflected back, the reflected light passes through the polishing slurry and the window, is received by the optical fiber, and transmitted to the photodetector to collect the energy reflectivity spectral information corresponding to the thickness of the polished layer.

[0033] The thickness of the polished layer during the polishing process is deduced based on the reflectance spectrum, mainly involving the establishment of a spectral library and library matching. Establishing the spectral library involves polishing a standard sample before polishing the product sample and collecting spectra corresponding to several discrete thicknesses for different polishing thicknesses, thus forming a spectral library. For example, offline thickness measurement is performed on the standard sample to determine its thickness distribution. After polishing, offline measurement is performed again to obtain the thickness distribution. Subsequently, each spectrum can be interpolated by region to assign thickness, thereby establishing a spectral library corresponding to different thicknesses.

[0034] Library matching includes: when grinding product wafers, comparing the real-time acquired spectrum with the spectrum in the spectral library, searching the spectral library (which can be referred to as library search), and finding the spectrum with the smallest mean squared error (MSE) with the acquired spectrum. This spectrum is considered to have the best match, and the thickness corresponding to the found spectrum is the thickness of the current grinding layer.

[0035] In related technologies, because the wear level of the grinding pads in grinding equipment varies daily, the spectral library needs to be updated daily or every other day to match the real-time working status of the CMP equipment. Furthermore, standard samples need to be ground before each update. In addition, since each grinding cycle removes a certain thickness of material, the collected spectra are actually from libraries with discrete thicknesses. There may also be cases where the wafer thickness is outside the library's range, so it's not always possible to find the optimal solution within the spectral library that perfectly matches the measured spectrum of the sample.

[0036] In view of this, embodiments of the present disclosure provide a method for flexibly establishing a spectral library that includes several theoretical energy reflectance spectra.

[0037] Figure 2 This is a flowchart illustrating a method for establishing a spectral library comprising several theoretical energy reflectance spectra according to an exemplary embodiment.

[0038] In this embodiment, the spectral library can be used to monitor the film thickness of the target sample during the CMP process. The spectral library may include several theoretical energy reflectance spectra. Film thickness monitoring includes acquiring the measured energy reflectance spectrum of the target sample through an optical detection module, and obtaining the current polishing layer thickness of the target sample by matching the measured energy reflectance spectrum with the spectral library. Film thickness monitoring can be used for end-point detection (EPD) of CMP; that is, if the current polishing layer thickness obtained from the spectral library does not reach the set polishing layer thickness, polishing continues; if the current polishing layer thickness obtained from the spectral library reaches the set polishing layer thickness, polishing stops, thereby achieving end-point monitoring of the polishing process.

[0039] like Figure 2 As shown, the method may include steps S21 to S24.

[0040] In step S21, the measured energy reflectance spectrum of the sample is acquired based on the optical detection module to obtain the first electric field reflectance and the second electric field reflectance.

[0041] Wherein, the first electric field reflectivity is the electric field reflectivity of the incident light from the optical detection module at the interface between the polishing slurry and the polishing layer of the target sample, and the second electric field reflectivity is the electric field emissivity of the incident light at the interface between the polishing layer of the target sample and the substrate layer of the target sample.

[0042] For ease of description, the polished layer of the target sample can be referred to as the first polished layer, and the base layer of the target sample can be referred to as the first base layer.

[0043] Figure 3 This is a schematic diagram illustrating the first and second electric field reflectivities according to an example.

[0044] In this embodiment, the first electric field reflectivity is the electric field reflectivity between the polishing slurry and the polishing layer. The composition of the polishing slurry used in establishing the spectral library and in actual measurements is the same. The composition of the polishing layer is the same for different wafers; for example, the material of the polishing layer can be SiO2, or other dielectric films can be selected as the polishing layer in this embodiment.

[0045] In this embodiment of the disclosure, the second electric field reflectivity is the electric field reflectivity between the polishing layer and the substrate layer (also known as the base layer) of the target sample.

[0046] In this embodiment of the disclosure, the first electric field reflectivity and the second electric field reflectivity can be derived from the collected measured energy reflectivity spectrum and theoretical calculations.

[0047] In this embodiment of the disclosure, the measured energy reflectance spectrum of the sample can be acquired by an optical detection module, and the first electric field reflectance and the second electric field reflectance can be derived by theoretical calculation based on the acquired measured energy reflectance spectrum.

[0048] In one example, a first measured energy reflectance spectrum of the calibration sample can be acquired using an optical detection module, and a first electric field reflectance can be calculated based on the first measured energy reflectance spectrum; the polishing slurry is introduced into the surface of the target sample, and a second measured energy reflectance spectrum of the target sample is acquired using an optical detection module, and a second electric field reflectance can be calculated based on the second measured energy reflectance spectrum and the first electric field reflectance.

[0049] In another example, the first measured energy reflectance spectrum of the calibration sample can be acquired by the optical detection module, and the first electric field reflectance can be calculated based on the first measured energy reflectance spectrum of the calibration sample; without the introduction of polishing slurry, the third measured energy reflectance spectrum of the target sample can be acquired by the optical detection module, and the second electric field reflectance can be calculated based on the third measured energy reflectance spectrum.

[0050] In another example, without introducing polishing slurry, the third measured energy reflectance spectrum of the target sample can be acquired by an optical detection module, and the second electric field reflectance can be calculated based on the third measured energy reflectance spectrum; when polishing slurry is introduced into the surface of the target sample, the second measured energy reflectance spectrum of the target sample can be acquired by an optical detection module, and the first electric field reflectance can be calculated based on the second measured energy reflectance spectrum and the second electric field reflectance.

[0051] In step S22, a theoretical total electric field reflectivity model is established based on the first electric field reflectivity, the second electric field reflectivity, the wavelength of the incident light, the refractive index of the incident light in the polishing layer, and the thickness of the polishing layer.

[0052] Optical endpoint detection technology used in CMP equipment is based on the principle of inversely calculating the film thickness on the sample surface using reflectance. In an EPD system, the emitted light from the light source is incident perpendicularly to the sample, and the light is directly reflected from the sample surface and received by the detector, thus acquiring the energy reflectance spectrum of the reflected light. The reflectance is related to the thickness of the polished layer, the refractive index, the wavelength of the incident light, and the incident angle. Therefore, through theoretical calculations, the film thickness of the polished sample can be inversely calculated based on the acquired reflectance spectrum.

[0053] Furthermore, since the sample is incident perpendicularly, the reflectivity of the beam reflected from the sample is the same in both the p and s directions, so only one direction needs to be calculated. Combined with... Figure 3 A simplified schematic diagram of the sample with a film structure, wherein the polishing slurry is layer 0, the material to be polished is layer 1 (i.e., the polishing layer), and the substrate layer is layer 2. Taking the calculation of the reflectivity of p-polarized light as an example, in this embodiment, the theoretical total electric field reflectivity model after the incident light is reflected by the target sample can be shown in formula (1):

[0054]

[0055] Where, r λ The theoretical total electric field reflectivity, r 01 R represents the reflectivity of the first electric field. 12 The second electric field reflectivity is represented by β = 2πnd / λ, where λ represents the wavelength of the incident light, n represents the refractive index corresponding to the incident light wavelength λ (n can be the complex refractive index of the SiO2 polishing layer), n is known, and d represents the thickness of the polishing layer.

[0056] Therefore, the first electric field reflectivity r is calculated based on the measured energy reflectivity spectrum of the sample. 01 Second electric field reflectivity r 12 Then, the theoretical total electric field reflectivity can be obtained through formula (1).

[0057] In step S23, based on the theoretical total electric field reflectivity model, the theoretical energy reflectivity model of the incident light after reflection by the target sample is obtained.

[0058] In this embodiment of the disclosure, the theoretical energy reflectivity model of the incident light after reflection from the target sample can be expressed as shown in formula (2):

[0059] R λ =|r λ | 2 (2)

[0060] Among them, R λ It represents energy reflectivity.

[0061] In this embodiment of the disclosure, after obtaining the theoretical total electric field reflectivity through formula (1), the theoretical energy reflectivity can be obtained through formula (2).

[0062] During EPD measurement, the signal actually acquired by the detector is the energy reflectivity R of the reflected light. λ (Also known as reflectance spectrum), that is, the spectrum obtained by EPD measurement is the energy reflectance R corresponding to the thickness of the polished layer film in the measurement area. λ Therefore, based on the first electric field reflectivity r 01 Second electric field reflectivity r 12 The precise value of the abrasive layer thickness can be used to calculate the theoretical energy reflectivity (also known as the theoretical spectrum) corresponding to any abrasive layer thickness, by combining formulas (1) and (2).

[0063] In step S24, based on the theoretical energy reflectance model, the theoretical energy reflectance spectra corresponding to different polishing layer thicknesses are calculated for use in film thickness monitoring, so as to generate a spectral library.

[0064] In this embodiment, the generated spectral library can be used during the actual grinding of the target sample to monitor the film thickness of the target sample. The spectral library may include theoretical energy reflectance spectra corresponding to different grinding layer thicknesses.

[0065] In this embodiment of the disclosure, the thickness step size and thickness range of the library can be set according to actual needs to generate a flexible and adjustable spectral library with different densities. For example, multiple grinding layer thicknesses can be set according to the range to which the grinding layer thickness of the actual target sample belongs, so that the set grinding layer thickness can cover the range to which the grinding layer thickness of the actual target sample belongs.

[0066] In an exemplary embodiment, a preset thickness step size can be adjusted to set multiple polishing layer thicknesses within a preset thickness range.

[0067] The preset thickness step size refers to the length between two adjacent grinding layer thicknesses, and the preset thickness range refers to the range between the maximum and minimum grinding layer thicknesses.

[0068] In this embodiment of the disclosure, the preset thickness step can be determined according to the required precision of the measurement, and the preset thickness range can be determined according to the range to which the thickness of the grinding layer of the actual grinding sample belongs.

[0069] Furthermore, by combining formulas (1) and (2), the theoretical energy reflectance spectra corresponding to different polishing layer thicknesses are calculated, thereby generating a spectral library. In an exemplary embodiment, the target sample has multiple feature regions, each feature region corresponding to a first electric field reflectance and a second electric field reflectance, and the spectral library includes several theoretical reflectance spectra corresponding to each feature region.

[0070] In this embodiment of the disclosure, the target sample can be a patterned sample, and the target sample can be divided into multiple feature regions based on its pattern characteristics. For example, the sample can be divided into a patterned area, an edge area, and a boundary area between the two. Data from each of the three feature regions is collected to obtain the first electric field reflectance and the second electric field reflectance corresponding to each feature region. Based on the first electric field reflectance and the second electric field reflectance corresponding to each feature region, a spectral library corresponding to each feature region is generated.

[0071] The method for establishing a spectral library comprising several theoretical energy reflectance spectra provided in this disclosure is based on the measured energy reflectance spectrum of a sample acquired by an optical detection module. It obtains the first electric field reflectance of incident light at the interface between the polishing slurry and the first polishing layer of the target sample, and the second electric field reflectance of incident light at the interface between the first polishing layer and the first substrate layer of the target sample. The theoretical energy reflectance corresponding to different polishing layer thicknesses can be determined using the first and second electric field reflectances to generate the spectral library. Since the first and second electric field reflectances can be accurately calculated by back-deriving from measured energy reflectance spectra of different samples and under different testing environments, the theoretical energy reflectance corresponding to different polishing layer thicknesses can be calculated based on the accurate values ​​of the first and second electric field reflectances. This generates a spectral library with flexible and adjustable data, improving the efficiency of establishing the spectral library. In other words, this invention can establish a theoretical energy reflectance spectral library corresponding to arbitrary thickness steps and thickness ranges, while obtaining a more refined spectral library and solving the problem of difficulty in solving theoretical models.

[0072] The following describes the specific process of obtaining the first electric field reflectance and the second electric field reflectance by acquiring the measured reflectance spectrum of the sample based on the optical detection module.

[0073] On the one hand, during the grinding process, the energy reflectivity R corresponding to several different grinding layer thicknesses in the same area is collected at different time points using an EPD system. λ From the spectrum, combined with formulas (1) and (2), the reflectivity r of the first electric field can be calculated in reverse. 01 Second electric field reflectivity r 12 This leads to the derivation of the theoretical spectral value corresponding to any polishing layer thickness. However, the first electric field reflectivity r 01 Second electric field reflectivity r 12 Since all values ​​are complex numbers, fitting solutions from multiple data points can lead to coupling issues, making it difficult to determine the first electric field reflectivity r. 01 Second electric field reflectivity r 12 The optimal solution is found. Therefore, embodiments of this disclosure provide a method for separately calculating the first electric field reflectivity r. 01 Second electric field reflectivity r12 The method.

[0074] On the other hand, the first electric field reflectivity r 01 Second electric field reflectivity r 12 All are related to the optical coefficients of the upper and lower layers of the interface, and satisfy the following relationship:

[0075]

[0076]

[0077] Where N2 represents the complex refractive index of the substrate, N0 represents the complex refractive index of the environment (in CMP, it is the complex refractive index of the polishing slurry), n is the complex refractive index of the polishing layer (e.g., SiO2), θ0 represents the incident angle of light in the polishing slurry, θ1 represents the incident angle of light at the interface between the polishing slurry and the polishing layer, and θ2 represents the incident angle of light at the interface between the polishing layer and the substrate, all of which are 0°.

[0078] Because polishing slurries have complex compositions and typically require on-site mixing, their refractive index N0 is difficult to calculate directly. Theoretical calculations are highly complex and impractical. Furthermore, the substrates of product samples are complex, often having undergone multiple photolithography, etching, and coating processes to form intricate patterns and structures (i.e., complex chip patterns), making it difficult to directly calculate the refractive index N2 of the substrate layer. Therefore, it is evident that, based on the existing conditions, it is difficult to solve for the first electric field reflectivity r. 01 Second electric field reflectivity r 12 In other words, it is impossible to obtain the theoretical energy reflectance spectrum of the sample through theoretical modeling.

[0079] In this embodiment of the disclosure, two intermediate variables are indirectly derived stepwise based on the spectra under different measurement conditions (including different spectral environmental conditions and / or different sample substrate conditions): the first electric field reflectance r at the interface between the polishing slurry and the first polishing layer. 01 and the second electric field reflectivity r at the interface between the polished layer and the first substrate layer 12 This allows us to deduce the energy reflectivity corresponding to different thicknesses of the polishing layer.

[0080] Figure 4 This is a flowchart illustrating, according to an exemplary embodiment, a method for obtaining a first electric field reflectance and a second electric field reflectance by acquiring the measured reflectance spectrum of a sample using an optical detection module. Figure 4 As shown, the method may include steps S41 to S48.

[0081] In step S41, a calibration sample is prepared.

[0082] The calibration sample includes a polished layer and a substrate layer. For ease of description, the polished layer of the calibration sample is referred to as the second polished layer, and the substrate layer of the calibration sample is referred to as the second substrate layer. The second polished layer of the calibration sample has a first complex refractive index, meaning the complex refractive index of the second polished layer is known. The material of the second polished layer of the calibration sample is the same as that of the first polished layer of the target sample; for example, the material of both polished layers can be SiO2, and the first complex refractive index is the complex refractive index of SiO2. The second substrate layer of the calibration sample has a second complex refractive index, meaning the complex refractive index of the second substrate layer is known; for example, the material of the second substrate layer can be Si, and the second complex refractive index is the complex refractive index of Si.

[0083] In step S42, the polishing slurry is introduced into the surface of the calibration sample, and the first measured energy reflectance spectrum at at least three measurement locations on the calibration sample is acquired based on the optical detection module.

[0084] In this embodiment of the present disclosure, the grinding slurry can be introduced into the surface of the calibration sample, and the first measured energy reflectance spectrum at at least three measurement positions on the calibration sample can be collected by the optical detection module, which facilitates subsequent regression fitting or simultaneous equations.

[0085] In this embodiment of the disclosure, the calibration sample can be a sample without a pattern, so when collecting the measured energy reflectance spectrum of the calibration sample, it is not necessary to distinguish the characteristic regions.

[0086] In step S43, the third electric field reflectivity of the incident light at the interface between the second polishing layer and the second substrate layer is determined based on the first complex refractive index and the second complex refractive index.

[0087] In this embodiment of the disclosure, the electric field reflectivity at the interface between the second polishing layer and the second substrate layer can be referred to as the third electric field reflectivity. The third electric field reflectivity between the second polishing layer and the second substrate layer of the calibration sample can be used... This indicates the reflectivity of the third electric field. The third electric field reflectivity can be calculated using formula (3). Substituting the first complex refractive index as the complex refractive index n of the polishing layer in formula (3) and the second complex refractive index as the complex refractive index N2 of the substrate layer in formula (3) into formula (3), the third electric field reflectivity can be calculated.

[0088] In step S44, the thickness of the second polishing layer at the measurement location on the calibration sample is measured.

[0089] In this embodiment of the disclosure, the thickness of the second polishing layer at at least three measurement locations on the calibration sample is measured using a film thickness measuring device. The first measured energy reflectance spectrum at each measurement location corresponds to the thickness of the polishing layer.

[0090] In step S45, the first electric field reflectivity is determined by solving simultaneous equations or regression fitting based on the third electric field reflectivity, the wavelength of the incident light, the first measured energy reflectivity spectrum, and the corresponding thickness of the second polishing layer.

[0091] In this embodiment of the disclosure, the first electric field reflectivity is a complex number, and can be expressed as r. 01 = a + bi. Where a represents the real part and b represents the imaginary part.

[0092] In the embodiments of this disclosure, based on the third electric field reflectivity, the wavelength of the incident light, the first measured energy reflectivity spectrum at the aforementioned at least three measurement locations, and the corresponding thickness of the second polishing layer, the third electric field reflectivity can be... r in formula (1) 12 Substituting into formula (1), according to formulas (1) and (2) and r 01 By solving the equations a + bi simultaneously, we can find a and b, thus determining the first electric field reflectivity.

[0093] In another embodiment of this disclosure, based on the third electric field reflectivity, the wavelength of the incident light, the first measured energy reflectivity spectrum at the aforementioned at least three measurement locations, and the corresponding thickness of the second polishing layer, the third electric field reflectivity can be... r in formula (1) 12 Substituting into formula (1), according to formulas (1) and (2) and r 01 =a + bi, through regression fitting, a and b are taken as variables to be solved, and the initial value is found using a grid-based regression algorithm (Newton's method, Levenberg-Marquardt (LM) algorithm, etc.). The objective function F1 that is minimized satisfies: F1 = ∑ x (R x -|r x | 2 ) 2 The subscript x indicates the number of data collection locations (greater than or equal to 3).

[0094] In this embodiment of the disclosure, the theoretical value of the first electric field reflectivity can be derived through the above steps.

[0095] In step S46, the polishing slurry is introduced into the surface of the target sample, and the second measured energy reflectance spectrum is acquired at at least three preset positions in the same feature area on the target sample based on the optical detection module.

[0096] In this embodiment of the disclosure, the target sample can also be referred to as a product sample. The material of the first polishing layer of the target sample is the same as the material of the second polishing layer of the calibration sample. The material of the first substrate layer of the target sample is more complex, therefore the complex refractive index of the first substrate layer of the target sample is unknown.

[0097] In this embodiment of the present disclosure, the polishing slurry can be introduced into the surface of the target sample, and the second measured energy reflectance spectrum at at least three preset positions in the same characteristic area on the target sample can be collected based on the optical detection module, which facilitates subsequent regression fitting or simultaneous equations.

[0098] In step S47, the thickness of the first polishing layer at a preset position on the target sample is measured.

[0099] In this embodiment of the disclosure, the thickness of the first polishing layer at at least three preset locations on the target sample is measured using a film thickness measuring device. The second measured energy reflectance spectrum at each preset location corresponds to the thickness of the polishing layer.

[0100] In step S48, the second electric field reflectivity is determined by solving simultaneous equations or regression fitting based on the first electric field reflectivity, the wavelength of the incident light, the second measured energy reflectivity spectrum, and the corresponding thickness of the first polishing layer.

[0101] In this embodiment of the disclosure, the second electric field reflectivity can be expressed as r 12 = m + ni. Where m represents the real part and n represents the imaginary part.

[0102] In the embodiments of this disclosure, the reflectivity of the first electric field, the wavelength of the incident light, the second measured energy reflectivity spectrum at the above-mentioned at least three preset positions, and the corresponding thickness of the first polishing layer can be determined based on formulas (1) and (2) and r. 12 By solving the equations m + ni simultaneously, we can find m and n, thus determining the second electric field reflectivity.

[0103] In another embodiment of this disclosure, the reflectivity of the first electric field, the wavelength of the incident light, the second measured energy reflectivity spectrum at the above-mentioned at least three preset positions, and the corresponding thickness of the first polishing layer can be determined based on formulas (1) and (2) and r. 12 =m+ni, through regression fitting, taking m and n as variables to be solved, using methods such as grid initialization + regression algorithm (Newton's method, Levenberg-Marquardt (LM) algorithm, etc.), the objective function F1 that is minimized satisfies: F1=∑ x (R x -|r x | 2 ) 2 The subscript x indicates the number of data collection locations (greater than or equal to 3).

[0104] The method disclosed herein for establishing a spectral library comprising several theoretical energy reflectance spectra combines measurement and theoretical calculation to solve for intermediate variables in the reflectance spectrum (i.e., the first electric field reflectance and the second electric field reflectance). Since the first electric field reflectance r... 01 This occurs at the interface between the polishing layer and the polishing slurry, and is independent of the substrate. Therefore, a calibration sample can be prepared by using a patternless pure silicon wafer as the substrate and depositing the layer to be polished. The third electric field reflectivity between the polishing layer and the substrate layer of the calibration sample can then be calculated. The energy reflectance spectrum of the calibration sample was collected and the thickness of the polished layer was measured. Then, based on the third electric field reflectance, the energy reflectance spectrum of the calibration sample, and the thickness of the polished layer, the first electric field reflectance r was calculated. 01 The first electric field reflectivity r was calculated. 01 Then, based on the first electric field reflectivity r 01 The second electric field reflectivity r is derived from the polishing layer thickness corresponding to the target sample and the second measured energy reflectivity spectrum. 12 Furthermore, based on the first electric field reflectivity r... 01 Second electric field reflectivity r 12 Calculate the theoretical energy reflectance spectra corresponding to different film thicknesses to generate a spectral library with flexible and adjustable data.

[0105] The following describes how the first electric field reflectivity r is determined in this embodiment. 01 Second electric field reflectivity r 12 The process will be explained in detail.

[0106] A calibration sample is prepared by using a bare die without patterns as a substrate and covering it with the material to be polished (e.g., SiO2). When the substrate is bare silicon, the third electric field reflectivity between the polishing layer and the substrate layer of the calibration sample under perpendicular incidence is directly calculated using formula (3) based on the complex refractive index of Si and the complex refractive index of SiO2.

[0107] The calibration sample was placed in the polishing slurry, and the reflected light energy and reflectivity at at least three locations within the same area on the surface of the calibration sample were collected using an EPD system. The thickness d at these locations was measured using other film thickness measuring equipment. t Therefore, the corresponding β can be calculated. t Value: β t =2πnd t / λ.

[0108] The electric field reflectivity at the interface between the slurry layer and the polishing layer of the calibration sample is set as the first electric field reflectivity r. 01 =a + bi. r 01It occurs at the interface between the Slurry and the polished silicon oxide layer, and is unrelated to the substrate.

[0109] On the one hand, based on the third electric field reflectivity The spectra and β values ​​at the aforementioned three or more positions of the calibrated sample were measured. t By solving the simultaneous equations, we can find a and b, and then obtain the first electric field reflectivity r at the interface between the polishing slurry and the polishing layer. 01 On the other hand, based on the third electric field reflectivity The spectra and β values ​​at the aforementioned three or more positions of the calibrated sample were measured. t Alternatively, it can be solved through regression fitting, with a and b as the variables to be solved. Initial values ​​can be found using methods such as grid-based methods and regression algorithms (Newton's method, LM algorithm, etc.). The objective function F1 that is minimized satisfies: F1 = ∑ x (R x -|r x | 2 ) 2 .

[0110] The target sample (also known as the product silicon wafer) is placed in a polishing slurry, and the spectra of at least three locations within the same area of ​​the target sample are acquired using an EPD system. λ The spectrum is preprocessed. The thickness d at these locations is measured using a film thickness measuring device, and the corresponding β is calculated from this, β = 2πnd / λ.

[0111] The electric field reflectivity, also known as the second electric field reflectivity r, is set at the interface between the substrate layer and the polishing layer of the target sample. 12 =m+ni.

[0112] On the one hand, based on the first electric field reflectivity r 01 (r 01 (a and b have been determined), the spectra and β at the aforementioned three or more positions of the target sample, and by solving the simultaneous equations, m and n can be obtained, and then the second electric field reflectivity r at the interface between the polished layer and the substrate of the target sample can be obtained. 12 On the other hand, based on the first electric field reflectivity r 01 (r 01 (Since a and b are already determined), the spectra and β at the aforementioned three or more positions of the target sample can also be obtained by regression fitting methods to solve for m and n. For example, initial value finding using a grid and regression algorithms (Newton's method, LM algorithm, etc.) can be employed to obtain m and n. The objective function F2 minimized in the regression algorithm satisfies: F2=∑ y (R y -|r y | 2 ) 2 .

[0113] Figure 5This is a flowchart illustrating, according to another exemplary embodiment, a method for obtaining a first electric field reflectance and a second electric field reflectance by acquiring the measured reflectance spectrum of a sample using an optical detection module. Figure 5 As shown, the method may include steps S51 to S59.

[0114] In step S51, a calibration sample is prepared.

[0115] In step S52, the polishing slurry is introduced into the surface of the calibration sample, and the first measured energy reflectance spectrum at at least three measurement locations on the calibration sample is acquired based on the optical detection module.

[0116] In step S53, the third electric field reflectivity of the incident light at the interface between the second polishing layer and the second substrate layer is determined based on the first complex refractive index and the second complex refractive index.

[0117] In step S54, the thickness of the second polishing layer at the measurement position on the calibration sample is measured.

[0118] In step S55, the first electric field reflectivity is determined by solving simultaneous equations or regression fitting based on the third electric field reflectivity, the wavelength of the incident light, the first measured energy reflectivity spectrum, and the corresponding thickness of the second polishing layer.

[0119] In this embodiment of the disclosure, the implementation process of steps S51 to S55 can refer to steps S41 to S45, and will not be repeated here.

[0120] In step S56, without introducing polishing slurry, the third measured energy reflectance spectrum is acquired at at least three preset positions within the same feature region on the target sample based on the optical detection module.

[0121] In this embodiment of the present disclosure, without introducing polishing slurry, the corresponding third measured energy reflectance spectra are collected at at least three preset locations within the same characteristic region on the target sample. That is, for each characteristic region of the target sample, the third measured energy reflectance spectra are collected at at least three preset locations to facilitate subsequent regression fitting or simultaneous equation solving.

[0122] In step S57, the fourth electric field reflectivity of the incident light at the interface between the air and the first polishing layer is determined based on the complex refractive index of air and the first complex refractive index.

[0123] In this embodiment of the disclosure, the electric field reflectivity at the interface between the first polishing layer and air can be referred to as the fourth electric field reflectivity. The fourth electric field reflectivity can be used... This indicates the reflectivity of the third electric field. The third electric field reflectivity can be calculated using formula (3). Substituting the first complex refractive index as the complex refractive index n of the polishing layer in formula (3) and the second complex refractive index as the complex refractive index N2 of the substrate layer in formula (3) into formula (3), the third electric field reflectivity can be calculated.

[0124] In this embodiment, the fourth electric field reflectivity between the first polishing layer and the air can be obtained by formula (3). Taking the first complex refractive index (e.g., the complex refractive index of SiO2) as n and the complex refractive index of air as N2, and substituting them into formula (3), the fourth electric field reflectivity between the first polishing layer and the air is obtained.

[0125] In step S58, the thickness of the first polishing layer at a preset position on the target sample is measured.

[0126] In step S59, the second electric field reflectivity is determined by solving simultaneous equations or regression fitting based on the fourth electric field reflectivity, the wavelength of the incident light, the third measured energy reflectivity spectrum, and the corresponding thickness of the first polishing layer.

[0127] In this embodiment of the disclosure, the second electric field reflectivity can be expressed as r 12 = m + ni. Where m represents the real part and n represents the imaginary part.

[0128] In one embodiment of this disclosure, based on the third electric field reflectivity, the wavelength of the incident light, the first measured energy reflectivity spectrum at the aforementioned at least three measurement locations, and the corresponding thickness of the second polishing layer, the third electric field reflectivity can be... r in formula (1) 12 Substituting into formula (1), according to formulas (1) and (2) and r 01 By solving the equations a + bi simultaneously, we can find m and b, thus determining the second electric field reflectivity.

[0129] In another embodiment of this disclosure, based on the fourth electric field reflectivity, the wavelength of the incident light, the third measured energy reflectivity spectrum at the aforementioned at least three measurement locations, and the corresponding thickness of the first polishing layer, the fourth electric field reflectivity can be... r in formula (1) 01 Substituting into formula (1), according to formulas (1) and (2) and r 12 =m+ni, through regression fitting, taking m and n as variables to be solved, using methods such as grid initialization and regression algorithms (Newton's method, Levenberg-Marquardt (LM) algorithm, etc.), the objective function F1 that is minimized satisfies: F1=∑ x (R x -|r x |2 ) 2 The subscript x indicates the number of data collection locations (greater than or equal to 3).

[0130] In this embodiment of the disclosure, the theoretical value of the second electric field reflectivity can be derived through the above steps.

[0131] The method disclosed herein for establishing a spectral library comprising several theoretical energy reflectance spectra combines measurement and theoretical calculation to solve for intermediate variables in the reflectance spectrum (i.e., the first electric field reflectance and the second electric field reflectance). Since the first electric field reflectance r... 01 This occurs at the interface between the polishing layer and the polishing slurry, and is independent of the substrate. Therefore, a calibration sample can be prepared by using a patternless pure silicon wafer as the substrate and depositing the layer to be polished. The third electric field reflectivity between the polishing layer and the substrate layer of the calibration sample can then be calculated. The energy reflectance spectrum of the calibration sample was collected and the thickness of the polished layer was measured. Then, based on the third electric field reflectance, the energy reflectance spectrum of the calibration sample, and the thickness of the polished layer, the first electric field reflectance r was calculated. 01 On the other hand, due to the second electric field reflectivity r 12 The location of the problem occurs at the interface between the substrate layer and the polishing layer of the target sample, and is independent of the polishing slurry. Therefore, the target sample is placed in air, and its energy reflectance spectrum is collected, while the thickness of the polishing layer is measured. Since the complex refractive indices of air and the polishing layer are known, the fourth electric field reflectance is calculated based on these indices. The second electric field reflectance r of the target sample can then be calculated independently using the fourth electric field reflectance, the energy reflectance spectrum of the target sample, and the thickness of the polishing layer. 12 Furthermore, based on the first electric field reflectivity r... 01 Second electric field reflectivity r 12 Calculate the theoretical energy reflectance spectra corresponding to different film thicknesses to generate a spectral library with flexible and adjustable data.

[0132] The following describes how r is determined in this embodiment. 01 and r 12 The process will be explained in detail.

[0133] A calibration sample (also known as the first sample) is prepared by using a bare silicon wafer without a pattern as the substrate and covering it with the material to be polished (e.g., SiO2). When the substrate is bare silicon, the third electric field reflectivity between the polishing layer and the substrate layer of the calibration sample under perpendicular incidence is directly calculated using formula (3) based on the complex refractive index of Si and the complex refractive index of SiO2.

[0134] The calibration sample was placed in the polishing slurry, and the reflected light energy and reflectivity at at least three locations within the same area on the surface of the calibration sample were collected using an EPD system. The thickness d at these locations was measured using other film thickness measuring equipment. t Therefore, the corresponding β can be calculated. t Value: β t =2πnd t / λ.

[0135] The electric field reflectivity at the interface between the slurry layer and the polishing layer when setting the calibration sample is also known as the first electric field reflectivity r. 01 =a + bi. r 01 It occurs at the interface between the Slurry and the polished silicon oxide layer, and is unrelated to the substrate.

[0136] On the one hand, based on the third electric field reflectivity The spectra and β values ​​at the aforementioned three or more positions of the calibrated sample were measured. t By solving the simultaneous equations, we can find a and b, and then obtain the first electric field reflectivity r at the interface between the polishing slurry and the polishing layer. 01 On the other hand, based on the third electric field reflectivity The spectra and β values ​​at the aforementioned three or more positions of the calibrated sample were measured. t Alternatively, it can be solved through regression fitting, with a and b as the variables to be solved. Initial values ​​can be found using methods such as grid-based methods and regression algorithms (Newton's method, LM algorithm, etc.). The objective function F1 that is minimized satisfies: F1 = ∑ x (R x -|r x | 2 ) 2 .

[0137] Before the polishing slurry is introduced, the spectral density (R) of the target sample (product silicon wafer) is collected at at least three locations within the same area using an EPD system. λ At this point, the environmental complex refractive index in formula (4) is the refractive index of air. Then, the spectrum is preprocessed, and the thickness d at these locations is measured using other film thickness measurement equipment. The corresponding β parameter is then calculated, β = 2πnd / λ.

[0138] Based on the refractive indices of air and the SiO2 polishing layer, the fourth electric field reflectivity is obtained using formula (4). The value of .

[0139] The electric field reflectivity at the interface between the substrate layer and the polishing layer is defined as the second electric field reflectivity r. 12 =m+ni. (r 12 The location of this event at the interface between the substrate and silica is independent of the polishing slurry, i.e., it is independent of whether CMP is performed.

[0140] On the one hand, based on the fourth electric field reflectivity By combining the spectra and β values ​​at the aforementioned three or more locations on the target sample with simultaneous equations, m and n can be calculated, and then the second electric field reflectivity r at the interface between the polishing layer and the substrate can be obtained. 12 On the other hand, based on the fourth electric field reflectivity The spectra and β values ​​at the aforementioned three or more positions of the target sample can also be obtained through regression fitting. m and n are the variables to be determined, and initial values ​​are obtained using methods such as grid-based initialization and regression algorithms (Newton's method, LM algorithm, etc.). The objective function F2 that is minimized satisfies: F2=∑ y (R y -|r y | 2 ) 2 .

[0141] Figure 6 This is a flowchart illustrating, according to yet another exemplary embodiment, a method for obtaining a first electric field reflectance and a second electric field reflectance by acquiring the measured reflectance spectrum of a sample using an optical detection module. Figure 6 As shown, the method may include steps S61 to S66.

[0142] In step S61, without introducing polishing slurry, the third measured energy reflectance spectrum is acquired at at least three preset positions within the same feature area on the target sample based on the optical detection module.

[0143] In step S62, the fourth electric field reflectivity of the incident light at the interface between the air and the first polishing layer is determined based on the complex refractive index of air and the first complex refractive index of the first polishing layer.

[0144] In step S63, the thickness of the first polishing layer at a preset position on the target sample is measured.

[0145] In step S64, the second electric field reflectivity is determined by solving simultaneous equations or regression fitting based on the fourth electric field reflectivity, the wavelength of the incident light, the third measured energy reflectivity spectrum, and the corresponding thickness of the first polishing layer.

[0146] In this embodiment of the disclosure, the implementation process of steps S61 to S64 can refer to steps S56 to S59, and will not be repeated here.

[0147] In step S65, the polishing slurry is introduced into the surface of the target sample, and the second measured energy reflectance spectrum is acquired at at least three preset positions in the same feature area on the target sample based on the optical detection module.

[0148] In this embodiment of the disclosure, the implementation process of step S65 can refer to step S46, and will not be repeated here.

[0149] In step S66, the first electric field reflectivity is determined by solving simultaneous equations or regression fitting based on the second electric field reflectivity, the wavelength of the incident light, the second measured energy reflectivity spectrum, and the corresponding thickness of the first polishing layer.

[0150] In this embodiment of the disclosure, the first electric field reflectivity is expressed as r. 01 = a + bi. Where a represents the real part and b represents the imaginary part.

[0151] In one embodiment of this disclosure, the second electric field reflectivity, the wavelength of the incident light, the second measured energy reflectivity spectrum at at least three preset positions, and the corresponding thickness of the first polishing layer can be used to determine the reflectivity based on formulas (1) and (2) and r. 01 By solving the equations a + bi simultaneously, we can find a and b, thus determining the first electric field reflectivity.

[0152] In another embodiment of this disclosure, the second electric field reflectivity, the wavelength of the incident light, the second measured energy reflectivity spectrum at the above-mentioned at least three preset positions, and the corresponding thickness of the first polishing layer can be used to determine the reflectivity based on formulas (1) and (2) and r. 01 =a + bi, through regression fitting, a and b are taken as variables to be solved, and the initial value is found using a grid-based regression algorithm (Newton's method, Levenberg-Marquardt (LM) algorithm, etc.). The objective function F1 that is minimized satisfies: F1 = ∑ x (R x -|r x | 2 ) 2 The subscript x indicates the number of data collection locations (greater than or equal to 3).

[0153] The method disclosed herein for establishing a spectral library comprising several theoretical energy reflectance spectra combines measurement and theoretical calculation to solve for intermediate variables in the reflectance spectrum (i.e., the first electric field reflectance and the second electric field reflectance). Since the second electric field reflectance r... 12 The location of the problem occurs at the interface between the substrate layer and the polishing layer of the target sample, and is independent of the polishing slurry. Therefore, the target sample is placed in air, and its energy reflectance spectrum is collected, while the thickness of the polishing layer is measured. Since the complex refractive indices of air and the polishing layer are known, the fourth electric field reflectance is calculated based on these indices. The second electric field reflectance r of the target sample can then be calculated independently using the fourth electric field reflectance, the energy reflectance spectrum of the target sample, and the thickness of the polishing layer. 12 The calculated reflectivity r of the second electric field is... 12 Then, based on the second electric field reflectivity r 12The first electric field reflectivity r is derived from the polishing layer thickness corresponding to the target sample and the second measured energy reflectivity spectrum. 01 Furthermore, based on the first electric field reflectivity r... 01 Second electric field reflectivity r 12 Calculate the theoretical energy reflectance spectra corresponding to different film thicknesses to generate a spectral library with flexible and adjustable data.

[0154] The following describes how r is determined in this embodiment. 01 and r 12 The process will be explained in detail.

[0155] Before the polishing slurry is introduced, the spectral density (R) of the target sample (product silicon wafer) is collected at at least three locations within the same area using an EPD system. λ At this point, the environmental complex refractive index in formula (4) is the refractive index of air. Then, the spectrum is preprocessed, and the thickness d at these locations is measured using other film thickness measurement equipment. The corresponding β parameter is then calculated, β = 2πnd / λ.

[0156] Based on the refractive indices of air and the SiO2 polishing layer, the fourth electric field reflectivity is obtained using formula (4). The value of .

[0157] The electric field reflectivity at the interface between the substrate layer and the polishing layer is defined as the second electric field reflectivity r. 12 =m+ni. (r 12 The location of this event at the interface between the substrate and silica is independent of the polishing slurry, i.e., it is independent of whether CMP is performed.

[0158] On the one hand, based on the fourth electric field reflectivity By combining the spectra and β values ​​at the aforementioned three or more locations on the target sample with simultaneous equations, m and n can be calculated, and then the second electric field reflectivity r at the interface between the polishing layer and the substrate can be obtained. 12 On the other hand, based on the fourth electric field reflectivity The spectra and β values ​​at the aforementioned three or more positions of the target sample can also be obtained through regression fitting. m and n are the variables to be determined, and initial values ​​are obtained using methods such as grid-based initialization and regression algorithms (Newton's method, LM algorithm, etc.). The objective function F2 that is minimized satisfies: F2=∑ y (R y -|r y | 2 ) 2 .

[0159] The target sample (product silicon wafer) was placed in polishing slurry, and the spectral R values ​​at at least three locations within the same area of ​​the target sample (product silicon wafer) were acquired using an EPD system. λThe spectrum is preprocessed, and the thickness d at these locations is measured using other film thickness measurement equipment. The corresponding β is then calculated, β = 2πnd / λ.

[0160] On the one hand, based on the second electric field reflectivity r obtained under the aforementioned condition without polishing fluid... 12 The values ​​of the spectrum and β at the aforementioned three or more locations on the target sample can be used to solve simultaneous equations to obtain m and n, and then the first electric field reflectivity r at the interface between the polishing slurry and the polishing layer can be obtained. 01 On the other hand, based on the second electric field reflectivity r obtained under the aforementioned condition without abrasive slurry... 12 The values, including the spectra and β values ​​at the aforementioned three or more positions of the target sample, can also be obtained through regression fitting. m and n are the variables to be determined, and are obtained using methods such as grid-based initial value finding and regression algorithms (Newton's method, LM algorithm, etc.). The objective function F3 that is minimized satisfies: F3 = ∑ z (R z -|r z | 2 ) 2 The subscript z indicates the data collection location.

[0161] In this embodiment, the steps of calculating the first electric field reflectivity and the second electric field reflectivity in different characteristic regions of the target sample are repeated in any of the above embodiments to calculate the first electric field reflectivity and the second electric field reflectivity corresponding to each wavelength in different characteristic regions of the target sample. For silicon wafers with complex chip patterns, characteristic regions can be divided according to the pattern features. For example, data from three regions can be collected: the pattern area, the edge area, and the boundary area between the two. Since the spectral measurement system uses a broadband light source, such as a light source in the 400 to 1000 nm band, which has 601 wavelength points, a total of 601 measured reflectivities R are obtained. λ The first and second electric field reflectivities corresponding to each wavelength in each feature region can be obtained through the above steps.

[0162] The thickness step size and thickness fluctuation range are preset. Based on the first electric field reflectivity and the second electric field reflectivity, as well as formulas (1) and (2), the theoretical spectrum of each feature region is obtained to establish the corresponding spectral library.

[0163] In this embodiment, a library with arbitrary thickness step size and thickness range can be established to obtain a more refined spectral library. When used for online measurement of sample film thickness in CMP equipment, the spectral library can be adjusted in real time according to the polished sample to find the optimal matching theoretical spectrum corresponding to the thickness of the polished layer film. Furthermore, regression methods can be used to further improve measurement accuracy after library matching.

[0164] Figure 7This is a flowchart illustrating a method for monitoring the thickness of an abrasive layer during chemical mechanical polishing, according to an exemplary embodiment. Figure 7 As shown, the method may include steps S71 to S73.

[0165] In step S71, during the grinding process, the current measured energy reflectance spectrum of the target sample is acquired in real time based on the optical detection module.

[0166] In an exemplary embodiment, the currently measured energy reflectance spectrum is the energy reflectance spectrum after preprocessing the initial spectral values ​​of the energy reflectance collected by the optical detection module. The spectral preprocessing may include background noise removal and filtering / smoothing.

[0167] In some embodiments, when the measurement window rotates out of the area covered by the carrier head, the spectrum at this point is acquired as the background noise spectrum. The energy reflectance spectrum of the bare silicon wafer under perpendicular incidence conditions is acquired, and the background noise spectrum is subtracted to obtain the reference spectrum for all types of feature regions.

[0168] In some embodiments, the spectra acquired in real time during the CMP process are corrected using the following formula:

[0169]

[0170] Where Is is the real-time sample light intensity, Isi is the bare silicon light intensity (after subtracting stray light), Id is the stray light intensity, and Rsi is the theoretical reflectivity of bare silicon, which can be calculated using equations (1) to (4). In actual measurements, the stray light intensity Id in the molecule can be updated in real time to eliminate the influence caused by changes in the polishing pad.

[0171] Since the EPD system acquires spectra in a single, high-speed manner, its signal-to-noise ratio cannot be improved by increasing the acquisition time. Therefore, the spectrum can be filtered and smoothed, such as by mean filtering or Fourier Transform (FFT) low-pass filtering. After processing, the signal-to-noise ratio of the spectrum can be significantly improved, thereby enhancing the accuracy of the analysis.

[0172] In step S72, a target theoretical energy reflectance spectrum that matches the current measured energy reflectance spectrum is determined from the spectral library.

[0173] The spectral library is determined according to any of the methods mentioned above.

[0174] In this embodiment of the disclosure, the preprocessed spectrum R can be used for library matching. When the accuracy requirement is not high and the library density is large enough, the search results can be used directly; alternatively, an initial value can be found first using library matching, and then the thickness d can be solved using a fitting regression method.

[0175] In this embodiment of the disclosure, the search range can be limited by the current number of rotations to improve speed.

[0176] In an exemplary embodiment, determining a target theoretical energy reflectance spectrum that matches the current measured energy reflectance spectrum from a spectral library includes: determining the current number of rotations of the bearing head used to grind the target sample, and the target feature region corresponding to the spectral acquisition position; determining the target grinding layer thickness range corresponding to the current number of rotations based on the correspondence between the number of rotations of the bearing head and the grinding layer thickness range; and determining the target theoretical energy reflectance spectrum that matches the current measured energy reflectance spectrum from the theoretical energy reflectance spectra in the spectral library that correspond to the target feature region and the target grinding layer thickness range.

[0177] In this embodiment, a correlation coefficient (e.g., goodness of fit (GOF)) filtering function can be set. Considering that some spectra will not be in the pre-defined regions, GOF calculation can be performed based on the theoretical and measured spectra after obtaining the thickness d. If the calculated GOF is less than a preset value (e.g., 0.9), the spectrum and thickness result are discarded. If the calculated GOF is greater than or equal to the preset value (e.g., 0.9), the spectrum and thickness result are determined to be usable.

[0178] In step S73, the thickness value corresponding to the target theoretical energy reflectance spectrum is determined as the thickness of the current polishing layer.

[0179] In this embodiment of the disclosure, the thickness of the polishing layer corresponding to the target theoretical energy reflectivity in the spectral library is the thickness of the current polishing layer.

[0180] In this embodiment of the present disclosure, if the current grinding layer thickness has not reached the set grinding layer thickness, grinding continues, and the current grinding layer thickness is determined by the above method after each grinding; when the current grinding layer thickness reaches the set grinding layer thickness, grinding stops, thereby realizing grinding terminal monitoring.

[0181] The grinding layer thickness detection method provided in this embodiment does not require updating the spectral library daily or every other day because the spectral library is generated based on the first electric field reflectivity and the second electric field reflectivity, which are independent of the wear degree of the grinding pad of the grinding equipment.

[0182] The method for monitoring the thickness of the abrasive layer during chemical mechanical polishing provided in this disclosure may include the following steps.

[0183] In some embodiments, real-time stray light calibration is performed. A large spot size (spot diameter > 1 mm) is used for acquisition, and stray light (i.e., reflected light from the polishing pad and polishing fluid) noise signal is acquired in each revolution and used to correct the measured spectrum to eliminate the influence of polishing pad surface variations; spectral smoothing (filtering, etc.) is also performed.

[0184] In some embodiments, a spectral library is established. Based on spectra collected under different measurement conditions (including different spectral environmental conditions and / or different sample substrate conditions), the electric field reflectivity at the interface between the grinding slurry and the grinding layer during the grinding process, as well as the electric field reflectivity at the interface between the grinding layer and the substrate, are indirectly deduced step by step. Then, the theoretical spectra corresponding to different film thicknesses are calculated to generate a theoretical spectral library.

[0185] In some embodiments, the same method is used to build theoretical libraries for different regions, depending on the actual situation of the silicon wafer (the structural characteristics of the patterned wafer surface).

[0186] In some embodiments, the measured data is processed and then matched against a library, with matching performed for each region. A fitting method can also be attempted.

[0187] In some embodiments, GOF filtering is added. Considering that not every acquired spectrum conforms to the above-mentioned library construction region, a theoretical curve and a measured curve are simulated based on the final thickness to calculate the correlation coefficient GOF. An appropriate value is set based on experience, such as 0.9. When the value is less than this, the result of the spectrum is discarded.

[0188] Based on the same concept, embodiments of this disclosure also provide an apparatus for monitoring the thickness of the polished layer on a sample during chemical mechanical polishing.

[0189] It is understood that the apparatus for monitoring the thickness of the polished layer of a sample during chemical mechanical polishing provided in this disclosure includes hardware structures and / or software modules corresponding to each function in order to achieve the above-mentioned functions. In conjunction with the units and algorithm steps of the various examples disclosed in this disclosure, this disclosure can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the technical solutions of this disclosure.

[0190] Figure 8 This is a block diagram illustrating an apparatus for monitoring the thickness of a polished layer on a sample during chemical mechanical polishing, according to an exemplary embodiment. (Refer to...) Figure 8 The device 800 includes an optical detection module 801 and a computing processing module 802.

[0191] The optical detection module 801 is used to acquire the current measured energy reflectance spectrum of the target sample in real time during the grinding process. The target sample includes the grinding layer and the substrate layer. The calculation and processing module 802 is used to establish a spectral library and determine the target theoretical energy reflectance spectrum that matches the current measured energy reflectance spectrum from the spectral library, and determine the thickness value corresponding to the target theoretical energy reflectance spectrum as the thickness of the current grinding layer.

[0192] In some embodiments, the calculation and processing module 802 is configured to: determine the current number of rotations of the bearing head used for grinding the target sample, and the target feature region corresponding to the current spectral acquisition position; determine the target grinding layer thickness range corresponding to the current number of rotations based on the correspondence between the number of rotations of the bearing head and the grinding layer thickness range; and determine the target theoretical energy reflectance spectrum that matches the current measured energy reflectance spectrum in the theoretical energy reflectance spectra corresponding to the target feature region and the target grinding layer thickness range in the spectral library.

[0193] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0194] This disclosure also provides a chemical mechanical polishing apparatus, including the aforementioned device for monitoring the thickness of the polished layer on a sample.

[0195] Although the operations are described in a specific order in the accompanying drawings, this should not be construed as requiring these operations to be performed in the specific order or serial order shown, or requiring all of the operations shown to obtain the desired result. In certain environments, multitasking and parallel processing may be advantageous.

[0196] The methods and apparatus disclosed herein can be implemented using standard programming techniques, utilizing rule-based logic or other logic to implement various method steps. It should also be noted that the terms "apparatus" and "module" as used herein and in the claims are intended to include implementations using one or more lines of software code and / or hardware implementations and / or devices for receiving input.

[0197] Any step, operation, or procedure described herein may be performed or implemented using one or more hardware or software modules, either alone or in combination with other devices. In one embodiment, the software module is implemented using a computer program product comprising a computer-readable medium containing computer program code, which is executable by a computer processor to perform any or all of the described steps, operations, or procedures.

[0198] The foregoing description of embodiments of this disclosure has been provided for purposes of illustration and description. The foregoing description is not exhaustive and is not intended to limit this disclosure to the exact form disclosed; various modifications and variations may be made in accordance with the foregoing teachings, or may be derived from practice of this disclosure. These embodiments were chosen and described to illustrate the principles of this disclosure and its practical application, enabling those skilled in the art to utilize this disclosure in various implementations and modifications suitable for the particular purpose conceived.

[0199] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0200] It is understood that in this disclosure, "multiple" refers to two or more, and other quantifiers are similar. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. The singular forms "a," "the," and "the" are also intended to include the plural forms unless the context clearly indicates otherwise.

[0201] It is further understood that the terms "first," "second," etc., are used to describe various types of information, but this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another, and do not indicate a specific order or degree of importance. In fact, the expressions "first," "second," etc., are completely interchangeable. For example, without departing from the scope of this disclosure, first information can also be referred to as second information, and similarly, second information can also be referred to as first information.

[0202] It can be further understood that, unless otherwise specified, "connection" includes both direct connections where no other components exist between the two parties and indirect connections where other components exist between them.

[0203] It is further understood that although operations are described in a specific order in the accompanying drawings in the embodiments of this disclosure, this should not be construed as requiring these operations to be performed in the specific order or serial order shown, or requiring all of the shown operations to be performed to obtain the desired result. In certain environments, multitasking and parallel processing may be advantageous.

[0204] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein.

[0205] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method of building a spectral library comprising a number of theoretical energy reflectance spectra, said spectral library being used for film thickness monitoring of a chemical mechanical polishing (CMP) process of a target wafer, said film thickness monitoring comprising acquisition of a measured energy reflectance spectrum of said target wafer by an optical probing module, characterized in that, The method comprises: Based on the optical detection module, the measured energy reflectance spectrum of the sample is collected, the first electric field reflectance of the incident light of the optical detection module at the interface between the grinding liquid and the first grinding layer of the target sample is obtained, and the second electric field reflectance of the incident light at the interface between the first grinding layer and the first substrate layer of the target sample is obtained; Based on the first electric field reflectance, the second electric field reflectance, the wavelength of the incident light, the refractive index of the incident light in the grinding layer, and the thickness of the grinding layer, a theoretical total electric field reflectance model of the incident light reflected by the target sample is established; Based on the theoretical total electric field reflectance model, a theoretical energy reflectance model of the incident light reflected by the target sample is obtained; Based on the theoretical energy reflectance model, the theoretical energy reflectance spectrum corresponding to different grinding layer thicknesses for the film thickness monitoring is calculated to generate the spectrum library.

2. The method of claim 1, wherein, Based on the optical detection module, the measured reflectance spectrum of the sample is collected, the first electric field reflectance and the second electric field reflectance are obtained, comprising: A calibration sample is prepared, the calibration sample comprising a second grinding layer having a first complex refractive index and a second substrate layer having a second complex refractive index, the second grinding layer being the same material as the first grinding layer of the target sample; The grinding liquid is introduced onto the surface of the calibration sample, and the first measured energy reflectance spectrum at at least three measurement positions on the calibration sample is collected based on the optical detection module; Based on the first complex refractive index and the second complex refractive index, a third electric field reflectance of the incident light at the interface between the second grinding layer and the second substrate layer is determined; The thickness of the second grinding layer at the measurement positions on the calibration sample is measured; Based on the third electric field reflectance, the wavelength of the incident light, the first measured energy reflectance spectrum, and the thickness of the second grinding layer, the first electric field reflectance is determined by simultaneous equations or regression fitting; The grinding liquid is introduced onto the surface of the target sample, and the second measured energy reflectance spectrum at at least three preset positions in the same feature area on the target sample is collected based on the optical detection module; The thickness of the first grinding layer at the preset positions on the target sample is measured; Based on the first electric field reflectance, the wavelength of the incident light, the second measured energy reflectance spectrum, and the thickness of the first grinding layer, the second electric field reflectance is determined by simultaneous equations or regression fitting.

3. The method of claim 1, wherein, Based on the optical detection module, the measured reflectance spectrum of the sample is collected, the first electric field reflectance and the second electric field reflectance are obtained, comprising: A calibration sample is prepared, the calibration sample comprising a second grinding layer having a first complex refractive index and a second substrate layer having a second complex refractive index, the second grinding layer being the same material as the first grinding layer of the target sample; The grinding liquid is introduced onto the surface of the calibration sample, and the first measured energy reflectance spectrum at at least three measurement positions on the calibration sample is collected based on the optical detection module; determining a third electric field reflectance of the incident light at an interface between the second polishing layer and the second substrate layer based on the first complex refractive index and the second complex refractive index; measuring respective thicknesses of the second polishing layer at the measurement positions on the calibration sample; determining the first electric field reflectance by simultaneous equations or regression fitting based on the third electric field reflectance, the wavelength of the incident light, the first measured energy reflectance spectrum and the respective thicknesses of the second polishing layer; collecting, by the optical detection module, at least three third measured energy reflectance spectra at the preset positions in the same feature region on the target sample without introducing the polishing liquid; determining a fourth electric field reflectance of the incident light at an interface between air and the first polishing layer based on a complex refractive index of air and the first complex refractive index of the first polishing layer; measuring respective thicknesses of the first polishing layer at the preset positions on the target sample; determining the second electric field reflectance by simultaneous equations or regression fitting based on the fourth electric field reflectance, the wavelength of the incident light, the third measured energy reflectance spectrum and the respective thicknesses of the first polishing layer.

4. The method of claim 1, wherein, The method for obtaining the first electric field reflectance and the second electric field reflectance based on the measured reflectance spectrum of the sample collected by the optical detection module comprises: collecting, by the optical detection module, at least three third measured energy reflectance spectra at the preset positions in the same feature region on the target sample without introducing the polishing liquid; determining a fourth electric field reflectance of the incident light at an interface between air and the first polishing layer based on a complex refractive index of air and the first complex refractive index of the first polishing layer; measuring respective thicknesses of the first polishing layer at the preset positions on the target sample; determining the second electric field reflectance by simultaneous equations or regression fitting based on the fourth electric field reflectance, the wavelength of the incident light, the third measured energy reflectance spectrum and the respective thicknesses of the first polishing layer. introducing the polishing liquid onto the surface of the target sample and collecting, by the optical detection module, at least three second measured energy reflectance spectra at the preset positions in the same feature region on the target sample; determining the first electric field reflectance by simultaneous equations or regression fitting based on the second electric field reflectance, the wavelength of the incident light, the second measured energy reflectance spectrum and the respective thicknesses of the first polishing layer.

5. The method according to any one of claims 1 to 4, characterized in that, Before the method calculates the theoretical energy reflectance spectrum corresponding to different polishing layer thicknesses for the film thickness monitoring call, the method further comprises: setting a plurality of polishing layer thicknesses in a preset thickness range with a preset thickness step.

6. The method according to any one of claims 1 to 4, characterized in that, The target sample has a plurality of feature regions, each of which corresponds to a first electric field reflectance and a second electric field reflectance, and the spectrum library comprises a plurality of theoretical reflectance spectra corresponding to each feature region.

7. A method for monitoring the thickness of a polishing layer during a chemical mechanical polishing process, characterized by, The method comprises: collecting, by the optical detection module, a current measured energy reflectance spectrum of the target sample in real time during the polishing process, wherein the target sample comprises a polishing layer and a substrate layer; determining a target theoretical energy reflectance spectrum matching the current measured energy reflectance spectrum from the spectrum library; determining a thickness value corresponding to the target theoretical energy reflectance spectrum as the current thickness of the polishing layer; wherein the spectrum library is established according to the method of any one of claims 1 to 6.

8. The method of claim 7, wherein, The current measured energy reflectance spectrum is an energy reflectance spectrum after background noise removal processing based on an initial energy reflectance spectrum collected by the optical detection module.

9. The method according to claim 7 or 8, characterized in that, The method of determining a target theoretical energy reflectance spectrum matching the current measured energy reflectance spectrum from the spectrum library comprises: determining a current rotation number of a carrier head used for polishing the target sample and a target feature area corresponding to a current spectrum collection position; determining a target polishing layer thickness range corresponding to the current rotation number based on a corresponding relationship between the rotation number of the carrier head and the polishing layer thickness range; determining a target theoretical energy reflectance spectrum matching the current measured energy reflectance spectrum from theoretical energy reflectance spectra corresponding to the target feature area and the target polishing layer thickness range in the spectrum library.

10. A device for monitoring the thickness of the polished layer on a sample during chemical mechanical polishing, characterized in that, The method comprises: an optical detection module for collecting a current measured energy reflectance spectrum of a target sample in real time during polishing, wherein the target sample comprises the polishing layer and a substrate layer; a calculation processing module for establishing a spectrum library and determining a target theoretical energy reflectance spectrum matching the current measured energy reflectance spectrum from the spectrum library, and determining a thickness value corresponding to the target theoretical energy reflectance spectrum as the current thickness of the polishing layer; wherein the spectrum library is determined according to the method of any one of claims 1 to 6.

11. A chemical mechanical polishing apparatus characterized by comprising: The device for monitoring the thickness of the polishing layer of the target sample of claim 10.

Images