Method for detecting polishing endpoint, chemical mechanical polishing apparatus, and storage medium

Abstract

The application provides a polishing endpoint detection method, a chemical mechanical polishing device and a storage medium, and belongs to the field of chemical mechanical polishing. The method comprises the following steps: acquiring a reflection spectrum of a wafer surface in a chemical mechanical polishing process; correcting the reflection spectrum by using a preset correction coefficient to obtain a corrected reflection spectrum; dividing the corrected reflection spectrum into a dark noise spectrum and a to-be-detected spectrum based on a comparison result of the corrected reflection spectrum and a corresponding threshold value; determining the reflectivity of the wafer surface based on the to-be-detected spectrum, the dark noise spectrum, a reference spectrum and a preset reference spectrum reflectivity; and determining the polishing endpoint based on the wafer thickness by using the reflectivity of the wafer surface to determine the wafer thickness. The application improves the accuracy of endpoint detection.

Description

Technical Field

[0001] This application belongs to the field of chemical mechanical polishing technology, and more specifically, relates to a polishing endpoint detection method, chemical mechanical polishing equipment, and storage medium. Background Technology

[0002] Chemical Mechanical Polishing (CMP) is an ultra-precision machining technique that achieves global planarization of wafer surfaces. Polishing endpoint detection is a crucial technology in CMP that determines when the polishing process ends, ensuring that the wafer achieves the preset flatness and material layer thickness. Insufficient accuracy in polishing endpoint detection can lead to under-polishing or over-polishing, thereby affecting wafer yield and production efficiency.

[0003] Currently, the industry mainly uses optical endpoint detection technology, which measures the intensity of reflected light on the polished surface and detects the polishing endpoint based on Fresnel's formula. However, in the actual polishing process, the measured reflected light intensity can be affected by measurement system errors, equipment errors, or environmental factors, resulting in deviations between the measured light intensity value and the theoretical light intensity value even for samples of the same thickness. This affects the accuracy of the endpoint detection results.

[0004] Therefore, improving the accuracy and reliability of CMP endpoint detection has become a key issue that the industry urgently needs to address. Summary of the Invention

[0005] The purpose of this application is to provide a polishing endpoint detection method, a chemical mechanical polishing device, and a storage medium, thereby solving or at least alleviating one or more of the above-mentioned and other problems existing in the prior art.

[0006] A first aspect of this application provides a polishing endpoint detection method, comprising:

[0007] Obtain the reflection spectrum of the wafer surface during chemical mechanical polishing;

[0008] The reflection spectrum is corrected using a preset correction factor to obtain the corrected reflection spectrum. The correction factor is calibrated based on the thickness change of the window at different wear stages in the polishing environment corresponding to the chemical mechanical polishing process.

[0009] Based on the comparison results between the corrected reflectance spectrum and the corresponding threshold, the corrected reflectance spectrum is divided into dark noise spectrum and the spectrum to be measured;

[0010] The reflectivity of the wafer surface is determined based on the spectrum to be measured, the dark noise spectrum, the reference spectrum, and the reflectivity of the preset reference spectrum.

[0011] The wafer thickness is determined by the reflectivity of the wafer surface, and the polishing endpoint is detected based on the wafer thickness.

[0012] In one embodiment, the method further includes: calibrating a correction coefficient based on the thickness variation of the window at different wear stages in a polishing environment corresponding to the chemical mechanical polishing process, further including:

[0013] Based on the polishing rate, polishing time, and initial thickness of the window, the theoretical thickness value of the window at different wear stages is determined;

[0014] Based on the initial thickness, initial reflection spectrum, and theoretical thickness value of the window, the theoretical reflection spectrum of the window at different wear stages is determined;

[0015] Obtain the actual reflectance spectra of the window at different wear stages;

[0016] The correction coefficients are determined based on the principle of minimizing the error between the theoretical reflectance spectrum and the corresponding actual reflectance spectrum.

[0017] In one embodiment, it further includes:

[0018] The calibration cycle for determining the correction coefficient is based on the light intensity attenuation of the light source.

[0019] Perform calibration operations on the correction coefficients according to the calibration cycle.

[0020] In one embodiment, performing a calibration operation on the correction coefficient includes:

[0021] The correction coefficient is calibrated based on the ratio of the initial light intensity to the current light intensity.

[0022] In one embodiment, the method further includes: determining a reference spectrum in a polishing environment corresponding to the chemical mechanical polishing process, including:

[0023] In a polishing environment corresponding to the chemical mechanical polishing process, a standard wafer corresponding to the wafer is chemically mechanically polished to obtain the reference reflectance spectrum of the standard wafer surface at different times during the polishing process;

[0024] The reference reflectance spectrum is corrected using a preset correction factor to obtain the corrected reference reflectance spectrum;

[0025] Determine the reference average light intensity and reference maximum light intensity of the corrected reference reflectance spectrum;

[0026] The reference average light intensity and the reference maximum light intensity are compared with the predetermined screening threshold, and the corrected reference reflectance spectrum whose comparison result is greater than the screening threshold is used as the screening spectrum.

[0027] The reference spectrum is determined based on the screened spectrum.

[0028] In one embodiment, based on the comparison result between the corrected reflectance spectrum and the corresponding threshold, the corrected reflectance spectrum is divided into a dark noise spectrum and a spectrum to be measured, including:

[0029] Determine the average light intensity, maximum light intensity, and spectral fit goodness of the corrected reflectance spectrum;

[0030] The corrected reflectance spectrum with maximum light intensity greater than the first threshold or average light intensity less than the second threshold, and the sampling point location is within the preset location range, is defined as the dark noise spectrum.

[0031] The corrected reflectance spectrum with an average light intensity or maximum light intensity greater than a predetermined screening threshold and a spectral goodness of fit greater than a preset goodness of fit threshold is determined as the spectrum to be measured.

[0032] In one embodiment, the method further includes: determining the average light intensity of all dark noise spectra to obtain a calibrated dark noise spectrum;

[0033] The reflectivity of the wafer surface is determined based on the measured spectrum, the dark noise spectrum, the reference spectrum, and the reflectivity of a preset reference spectrum, including:

[0034] The reflectivity of the wafer surface is determined based on the spectrum to be measured, the calibrated dark noise spectrum, the reference spectrum, and the reflectivity of the preset reference spectrum.

[0035] In one embodiment, determining the reflectivity of the wafer surface based on the measured spectrum, the calibrated dark noise spectrum, the reference spectrum, and the reflectivity of a preset reference spectrum includes:

[0036] The calibration dark noise spectrum is removed from the spectrum to be measured to obtain the first spectrum;

[0037] The second spectrum is obtained by removing the calibration dark noise spectrum from the reference spectrum;

[0038] The reflectance of the wafer surface is determined based on the first spectrum, the second spectrum, and the reflectance of a preset reference spectrum.

[0039] A second aspect of this application provides a chemical mechanical polishing apparatus, comprising: a polishing platform, a polishing pad, a polishing head, and an endpoint detection system; the endpoint detection system includes a light source, a spectral receiver, and a controller.

[0040] A polishing head is used to load a wafer and place it against a polishing pad above a polishing platform.

[0041] A light source, used to output light signals to the wafer surface;

[0042] A spectral receiver for receiving the spectrum reflected from the wafer surface;

[0043] The controller is used to perform the steps of the above method during the chemical mechanical polishing process on the wafer.

[0044] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0045] The beneficial effects of the polishing endpoint detection method, chemical mechanical polishing equipment, and storage medium provided in this application are as follows:

[0046] This application obtains the reflection spectrum of the wafer surface during chemical mechanical polishing, corrects the reflection spectrum using a preset correction coefficient, and obtains a corrected reflection spectrum. Based on the comparison between the corrected reflection spectrum and the corresponding threshold, the corrected reflection spectrum is divided into a dark noise spectrum and a test spectrum. The wafer thickness is then determined based on the test spectrum, the dark noise spectrum, the reference spectrum, and the reflectance of the preset reference spectrum. The polishing endpoint is then detected based on the wafer thickness.

[0047] This application corrects the obtained reflection spectrum using a preset correction coefficient to obtain a corrected reflection spectrum. The endpoint detection process uses this corrected reflection spectrum, which reduces measurement deviations caused by measurement system errors, equipment errors, or environmental factors. This makes the reflection spectrum used for endpoint detection closer to the theoretical value, thereby improving the accuracy of endpoint detection, avoiding over-polishing or under-polishing, and further improving the yield and polishing accuracy of wafer production.

[0048] The calibration process of the correction coefficient in this application is performed in a polishing environment corresponding to the chemical mechanical polishing process. This ensures that the spectrum used in the calibration process and the spectrum used for the end-point detection of the actual polishing process are the spectra under the corresponding environment. This effectively avoids the problem of inaccurate calibration of the correction coefficient due to environmental differences, improves the accuracy of the spectrum after correction based on the correction coefficient, thereby improving the accuracy and reliability of the end-point detection, avoiding over-polishing or under-polishing, and further improving the yield and polishing precision of wafer production. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a schematic diagram of a chemical mechanical polishing apparatus provided in an embodiment of this application;

[0051] Figure 2 A schematic flowchart of a polishing endpoint detection method provided in an embodiment of this application;

[0052] Figure 3 A simplified diagram of the measurement optical path used in the polishing endpoint detection method according to an embodiment of this application is provided;

[0053] Figure 4 A flowchart of a correction coefficient calibration method provided in an embodiment of this application;

[0054] Figure 5 A flowchart for dividing a modified reflectance spectrum into a dark noise spectrum and a spectrum to be measured, provided as an embodiment of this application;

[0055] Figure 6 A light intensity variation curve provided in an embodiment of this application;

[0056] Figure 7 A flowchart for determining a reference spectrum provided in one embodiment of this application;

[0057] Figure 8 The spectral fit goodness curve is shown using the existing polishing endpoint detection method.

[0058] Figure 9 The spectral fit goodness curve of the polishing endpoint detection method of this application is shown.

[0059] Figure 10 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0060] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0062] refer to Figure 1 This is a schematic diagram of the structure of the chemical mechanical polishing apparatus 100 provided in an embodiment of this application. Figure 1As shown, the chemical mechanical polishing (CMP) apparatus 100 mainly includes: a polishing head 102 for loading and rotating a wafer, a polishing platform 104 covered with a polishing pad 106, and a liquid supply device 108 for supplying polishing slurry. During CMP, the polishing head 102 can load the wafer and place it against the polishing pad 106 above the polishing platform 104, causing the wafer to rotate relative to the polishing platform 104 (refer to indicator arrow F1) and reciprocate along the radial direction of the polishing platform 104 (refer to indicator arrow F2). The polishing pad 106 also rotates simultaneously (refer to indicator arrow F3). The liquid supply device 108 supplies polishing slurry between the polishing pad 106 and the wafer. Under the action of the polishing slurry, the polishing pad 106 rubs the surface of the wafer, performing a CMP operation on the wafer. The chemical mechanical polishing apparatus 100 also includes an endpoint detection system (not shown in the figure), which includes a light source, a spectral receiver, and a controller. The light source is used to output a light signal that can reach the wafer surface through a window in the polishing pad 106. The spectral receiver can receive the light signal reflected from the wafer surface. The controller is used to control the chemical mechanical polishing apparatus 100 to perform the polishing endpoint detection method of any embodiment of this application based on the light signal during the chemical mechanical polishing of the wafer.

[0063] refer to Figure 2 , Figure 2 The flowchart illustrating a polishing endpoint detection method according to an embodiment of this application is shown. This method can be executed by a chemical mechanical polishing (CMP) apparatus 100, specifically by an endpoint detection system within the CMP apparatus 100. The polishing endpoint detection method of this application may include the following steps:

[0064] S11: Obtain the reflection spectrum of the wafer surface during chemical mechanical polishing.

[0065] In one embodiment, the polishing endpoint detection method of this application uses a simplified measurement optical path diagram reference. Figure 3 As shown, light emitted from the light source passes through the probe and enters the window of the polishing pad, reaching the wafer surface. It is then reflected back from the wafer surface to the window of the polishing pad and the probe, and finally received by a spectral receiver. The probe can be an optical sensor.

[0066] In one embodiment, the reflectance spectrum of the wafer surface can be acquired at different times during the chemical mechanical polishing process, thus enabling the acquisition of multiple frames of reflectance spectrum.

[0067] S12: Correct the reflection spectrum using a preset correction coefficient to obtain the corrected reflection spectrum.

[0068] To overcome the problem that the collected reflectance spectrum deviates from the theoretical value due to measurement system errors, equipment errors, or environmental factors during the polishing process, this application uses a preset correction coefficient k to correct the collected reflectance spectrum, thus obtaining a corrected reflectance spectrum. In other words, a correction operation is performed on the reflectance spectrum of each frame collected to obtain the corrected reflectance spectrum.

[0069] In one embodiment, the correction factor k is calibrated based on the thickness change of the window at different wear stages in a polishing environment corresponding to the chemical mechanical polishing process. That is, the polishing endpoint detection method of this application also includes the step of calibrating the correction factor.

[0070] The calibration process of the correction factor in this application is performed in a polishing environment corresponding to the chemical mechanical polishing process. The polishing environment corresponding to the chemical mechanical polishing process means that the polishing environment of the calibration process is as similar as possible to the actual chemical mechanical polishing process. The polishing environment includes: polishing consumables (such as polishing pads, polishing fluids, dressing heads and dressing discs, etc.) and light intensity measurement system, etc.

[0071] The spectrum used in the calibration process of the correction coefficient in this application is the spectrum under the corresponding environment as the spectrum used in the actual polishing process for the polishing endpoint detection. This can effectively avoid the problem of inaccurate calibration of the correction coefficient due to differences in the polishing environment, and can improve the accuracy of the spectrum after correction based on the correction coefficient, thereby improving the accuracy and reliability of the endpoint detection, avoiding over-polishing or under-polishing, and further improving the yield and polishing precision of wafer production.

[0072] During CMP (Continuous Polishing), the thickness of the window plate decreases due to wear. As the window plate gradually thins, the path of light propagation within it shortens, reducing light loss and thus increasing the intensity of the reflected spectrum received by the spectral receiver. This application utilizes the thickness variation of the window plate at different wear stages to obtain the theoretical and actual thickness values ​​of the window plate at different polishing moments. This allows for the calibration of the correction coefficient k based on multiple sets of theoretical and actual thickness values.

[0073] In one embodiment, a process reference for calibrating correction coefficients based on the thickness variation of the window at different wear stages is used in a polishing environment corresponding to a chemical mechanical polishing process. Figure 4 As shown, it includes:

[0074] S401. Based on the polishing rate, polishing time, and initial thickness of the window, determine the theoretical thickness value of the window at different wear stages.

[0075] In practice, the probe lens extends into the window, therefore the thickness of the window represents the distance between the probe and the wafer surface; the initial thickness of the window represents the initial distance between the probe and the wafer surface before polishing, as shown in the reference. Figure 3 As shown, the initial thickness of the window is .

[0076] If the polishing rate is r, then the theoretical thickness of the window at time t is... , It can be represented as:

[0077]

[0078] The different wear stages, i.e., the t values, in this application are different. Figure 3 d t This represents the decrease in the thickness of the window at time t.

[0079] S402. Based on the initial thickness of the window, the initial reflection spectrum, and the theoretical thickness value, determine the theoretical reflection spectrum of the window at different wear stages.

[0080] The initial reflection spectrum is the light intensity value of the reflection spectrum returned by the wafer surface in its initial state before polishing; the theoretical reflection spectrum of the window at time t is the light intensity value of the theoretical reflection spectrum returned by the wafer surface at time t, which can be expressed as:

[0081]

[0082] in, This represents the light intensity value of the initial reflected spectrum. This represents the light intensity value of the theoretical reflected spectrum at time t.

[0083] S403. Obtain the actual reflection spectrum of the window at different wear stages.

[0084] The actual reflectance spectrum at time t can be expressed as the spectrum obtained after correcting the theoretical reflectance spectrum at time t, specifically as follows:

[0085]

[0086] k is the correction factor. This represents the intensity value of the actual reflected spectrum after correction.

[0087] S404. Calibrate the correction coefficients based on the principle of minimizing the error between the theoretical reflection spectrum and the corresponding actual reflection spectrum.

[0088] To obtain the optimal correction coefficient, the error between the corrected actual reflectance spectrum and the theoretical reflectance spectrum should be minimized, i.e., let

[0089] In one embodiment, the correction coefficient is calibrated based on the principle of minimizing the error between the theoretical reflectance spectrum and the corresponding actual reflectance spectrum, including:

[0090] S4041. Obtain the sum of squares of the errors between the theoretical and actual reflectance spectra of the window at different wear stages. .

[0091]

[0092] n represents the number of experimental data sets obtained, I i and h i These are the light intensity and thickness values ​​of the actual reflected spectrum of the i-th set of data obtained in the experiment.

[0093] S4042. Differentiate and minimize the sum of squared errors to obtain the correction coefficients.

[0094] To obtain the optimal correction coefficient, it is necessary to... The optimal solution is obtained when the derivative is 0. That is, the derivative is set to 0, and the k obtained is the calibration correction coefficient.

[0095]

[0096] Then k can be expressed as

[0097] This application uses the correction coefficients calibrated by the above method to correct the reflectance spectrum, and the corrected reflectance spectrum can be expressed as:

[0098]

[0099] In one embodiment, after polishing for a period of time, the correction coefficient may drift due to changes in factors such as the polishing environment. For example, the light intensity of the light source may decrease after a period of time, causing the correction coefficient to drift. Therefore, this embodiment of the application also includes an operation of calibrating the correction coefficient.

[0100] In one embodiment, taking the drift of the correction coefficient caused by the attenuation of the light intensity of the light source as an example, the calibration period of the correction coefficient can be determined based on the attenuation of the light intensity of the light source, and the calibration operation can be performed on the correction coefficient according to the calibration period. For example, a correction coefficient calibration operation is performed once for every preset percentage attenuation of light intensity. This preset percentage can be, for example, 15%, 20%, or 25%, etc., and can be set as needed. Taking a preset percentage of 20% as an example, a correction coefficient calibration operation is performed once when the light intensity attenuates by 20%, 40%, 60%, etc. In another embodiment, the percentage of light intensity attenuation can be determined based on the number of times the light source flickers. For example, if the light source can flicker 1 billion times within its lifespan, the light intensity attenuation of the light source can be determined to be 20% when the light source flickers 200 million times, at which point a correction coefficient calibration operation can be performed.

[0101] In one embodiment, the method for calibrating the correction coefficient includes: re-performing the correction coefficient calibration operation using the current light source with the method described in the above embodiment, and using the recalibrated correction coefficient as the calibrated correction coefficient.

[0102] In another embodiment, the method for calibrating the correction coefficient includes calibrating the correction coefficient based on the ratio of the initial light intensity of the light source to the current light intensity. For example, the initial light intensity of the light source is... The initial calibration correction factor is The current light intensity is Then the correction factor after calibration It can be represented as:

[0103]

[0104] This application embodiment takes into account that the change in light source intensity will cause the correction coefficient to drift, and will periodically perform a calibration operation on the correction coefficient to ensure the accuracy of the correction coefficient, and further ensure the accuracy of the reflection spectrum after correction based on the correction coefficient, thereby ensuring the accuracy and reliability of the endpoint detection and improving the wafer processing yield.

[0105] S13: Based on the comparison results between the corrected reflectance spectrum and the corresponding threshold, the corrected reflectance spectrum is divided into dark noise spectrum and the spectrum to be measured.

[0106] This step distinguishes whether the corrected reflectance spectrum comes from the wafer or from dark noise values ​​by comparing the corrected reflectance spectrum with the corresponding threshold.

[0107] In one embodiment, the process of dividing the modified reflectance spectrum into a dark noise spectrum and the spectrum to be measured based on the comparison result between the modified reflectance spectrum and the corresponding threshold is as follows: Figure 5 As shown, it includes:

[0108] S501. Determine the average light intensity, maximum light intensity, and spectral fit goodness of the corrected reflection spectrum.

[0109] For the current time i, the corrected reflection spectrum of the wafer surface can be obtained, and its average light intensity can be calculated. Maximum light intensity and spectral fit goodness In this application embodiment, the average light intensity refers to the average light intensity of multiple bands included in the modified reflection spectrum; the maximum light intensity refers to the maximum light intensity of multiple bands included in the modified reflection spectrum.

[0110] S502. Compare the average light intensity, maximum light intensity, and spectral fit with the corresponding thresholds.

[0111] In one embodiment, corresponding thresholds can be set for the average light intensity, the maximum light intensity, and the spectral fit goodness, and the set thresholds include: a first threshold. Second threshold Filtering threshold and goodness-of-fit threshold Among them, the maximum light intensity is compared with the first threshold. and filtering threshold Compare; compare the average light intensity with the second threshold. and filtering threshold Compare the spectral goodness of fit with the goodness-of-fit threshold. The comparisons were made; based on the comparison results, the dark noise spectrum in the corrected reflectance spectrum and the spectrum to be measured from the wafer were separated.

[0112] S503. The corrected reflectance spectrum with the maximum light intensity greater than the first threshold or the average light intensity less than the second threshold, and the sampling point location within the preset location range, is determined as the dark noise spectrum.

[0113] This embodiment will satisfy and satisfy <Index< The corrected reflectance spectrum was determined to be the dark noise spectrum, and Index represents the location of the acquisition point. These represent the lower and upper limits of the range of the data collection point location, respectively.

[0114] In one embodiment, during chemical mechanical polishing, the acquisition point position of the obtained reflectance spectrum can be determined in real time based on parameters such as the oscillation period of the polishing platform 104. When determining the dark noise spectrum, in addition to comparing the maximum light intensity and average light intensity with the corresponding threshold, it is also determined whether it is a dark noise spectrum based on whether the acquisition point position is within a preset position range. If the acquisition point position is within the range of the acquisition point position set by the user, then it is determined that the following conditions are met simultaneously: In this case, the corrected reflectance spectrum can be confirmed as a dark noise spectrum. This application does not specifically limit the method for determining the location of the reflectance spectrum acquisition point; it can be implemented based on relevant technologies.

[0115] This embodiment effectively eliminates noise interference by filtering out the dark noise spectrum in the corrected reflectance spectrum, thereby improving the accuracy of endpoint detection.

[0116] In one embodiment, the method further includes: performing a calibration operation on the dark noise spectrum to obtain a calibrated dark noise spectrum. Specifically, the average light intensity of all dark noise spectra can be determined, and the average light intensity of all dark noise spectra can be used as the calibrated dark noise spectrum. The calibration operation on the dark noise spectrum can be expressed as follows:

[0117]

[0118] in, To calibrate the intensity values ​​of the dark noise spectrum, m represents the number of dark noise spectra. The light intensity value of the i-th dark noise spectrum in m groups of dark noise spectra.

[0119] The embodiments of this application can further improve the accuracy of denoising the corrected reflectance spectrum by performing a calibration operation on the dark noise spectrum, thereby improving the accuracy of endpoint detection.

[0120] S504. The corrected reflectance spectrum with an average light intensity or maximum light intensity greater than a predetermined screening threshold and a spectral fit greater than a preset fit threshold is determined as the spectrum to be tested.

[0121] This embodiment will satisfy The corrected reflection spectrum is determined as the spectrum to be measured, which is the reflection spectrum from the wafer surface.

[0122] like Figure 6 The graph shown is a light intensity variation curve, with the horizontal axis representing the sampling point and the vertical axis representing the light intensity value. The method described in this application can distinguish between the target spectrum and the dark noise spectrum on the wafer, thereby eliminating noise interference in the reflection spectrum and improving the accuracy of endpoint detection. Figure 6 As shown, the corrected reflectance spectrum includes not only the dark noise spectrum and the spectrum to be measured, but also other spectra that do not meet the judgment conditions for either the dark noise spectrum or the spectrum to be measured. These spectra can be ignored when performing polishing endpoint detection. In other words, the polishing endpoint detection process of this application does not use other spectra for calculation. Therefore, the method for distinguishing between the dark noise spectrum and the spectrum to be measured in this application can help to effectively filter out noise interference, thereby improving the accuracy of endpoint detection.

[0123] S14: Determine the reflectivity of the wafer surface based on the spectrum to be measured, the dark noise spectrum, the reference spectrum, and the reflectivity of the preset reference spectrum.

[0124] Existing polishing endpoint detection methods typically rely on statically acquired reference spectra and dark noise light as theoretical references when performing spectral analysis. However, in actual CMP processes, factors such as changes in polishing slurry composition and environmental interference can significantly affect the stability of spectral signals, leading to large deviations in polishing endpoint detection based on statically acquired reference spectra and dark noise light. This affects the accuracy of polishing endpoint detection, thereby impacting wafer yield and polishing precision.

[0125] To overcome the above problems, the embodiments of this application also include a step of calibrating the reference spectrum (i.e. calibrating the light intensity). Before performing the chemical mechanical polishing in step S11, the reference spectrum is dynamically determined based on the modified reflection spectrum in the polishing environment corresponding to the chemical mechanical polishing process.

[0126] In one embodiment, the process for determining the reference spectrum in the polishing environment corresponding to the chemical mechanical polishing process is as follows: Figure 7 As shown, it includes:

[0127] S701. Polish a standard wafer corresponding to the chemical mechanical polishing process in a polishing environment, and obtain the reference reflection spectrum of the surface of the standard wafer at different times during the polishing process.

[0128] In one embodiment, the standard wafer is a silicon wafer corresponding to the wafer in step S11. A chemical mechanical polishing process is performed on the standard wafer, and the reflection spectrum at different times during the polishing process is collected. The collected reflection spectrum is then used to calibrate a reference spectrum. To distinguish it from the reflection spectrum obtained during the actual wafer polishing process, this embodiment refers to the reflection spectrum obtained during the chemical mechanical polishing of the standard wafer as the reference reflection spectrum.

[0129] The environment in which the standard wafer is subjected to chemical mechanical polishing corresponds to, and is as consistent as possible with, the polishing environment in which the wafer is actually polished in step S11, in order to ensure the accuracy of the calibrated reference spectrum and improve the spectral fitting goodness.

[0130] S702. The reference reflectance spectrum is corrected using a preset correction coefficient to obtain the corrected reference reflectance spectrum.

[0131] Step S702 is the same as step S12, and will not be repeated here.

[0132] S703. Determine the reference average light intensity and reference maximum light intensity of the corrected reference reflection spectrum.

[0133] For the current time i, the corrected reference reflection spectrum of the wafer surface can be obtained, and its reference average light intensity and reference maximum light intensity can be calculated. The reference average light intensity is the average light intensity of multiple bands included in the corrected reference reflection spectrum; the reference maximum light intensity is the maximum light intensity of multiple bands included in the corrected reference reflection spectrum.

[0134] S704. Determine if CMP has ended.

[0135] If CMP has not ended, repeat steps S701 to S703 above. That is, during CMP, the reference average light intensity and reference maximum light intensity at different times are collected and recorded in real time, so that multiple frames of reference average light intensity and reference maximum light intensity can be obtained.

[0136] If CMP ends, proceed to step S705.

[0137] S705. Compare the reference average light intensity and the reference maximum light intensity with the predetermined screening thresholds, respectively.

[0138] Will and Each with a predetermined screening threshold Comparison, that is, judging whether the conditions are met. . and These represent the reference average light intensity and the reference maximum light intensity, respectively. This indicates the filtering threshold.

[0139] S706. Use the corrected reference reflectance spectrum whose comparison result is greater than the screening threshold as the screening spectrum.

[0140] That is, to satisfy or The corrected reference reflectance spectrum is used as the screening spectrum. Among them, for each frame and By performing the above judgment operation, multiple frames of screening spectra can be obtained.

[0141] S707. Determine the reference spectrum based on the screened spectrum.

[0142] In one embodiment, determining the reference spectrum based on the screening spectrum involves calculating the average light intensity value of multiple frames of screening spectra, which can be expressed as:

[0143]

[0144] in, To screen the light intensity values ​​of the spectrum, The light intensity value is the reference spectrum. Indicates the first The frame is used to filter the spectrum, where s represents the number of spectra to be filtered.

[0145] This application can dynamically determine the reference spectrum based on the actual polishing environment of the wafer, so that the spectrum used in the reference spectrum calibration process and the spectrum used for polishing endpoint detection in the actual polishing process are the spectra under the corresponding environment. This can avoid the problem of inaccurate reference spectrum calibration due to environmental differences, effectively improve the calculation accuracy, further ensure the polishing endpoint detection accuracy, and thus ensure the yield and polishing accuracy of wafer production.

[0146] In one embodiment, a method for determining the reflectivity of a wafer surface based on the measured spectrum, the dark noise spectrum, the reference spectrum, and the reflectivity of a preset reference spectrum includes:

[0147] First, the dark noise spectrum of the spectrum to be measured is removed to obtain the first spectrum, and the dark noise spectrum of the reference spectrum is removed to obtain the second spectrum. It can be understood that if a calibration operation is performed on the dark noise spectrum, the calibration dark noise spectrum of the spectrum to be measured is removed to obtain the first spectrum, and the calibration dark noise spectrum of the reference spectrum is removed to obtain the second spectrum.

[0148] Finally, based on the reflectance of the first spectrum, the second spectrum, and the preset reference spectrum, the reflectance of the wafer surface is determined, which can be expressed as:

[0149]

[0150] in, Indicates the reflectivity of the wafer surface. This represents the reference spectral reflectance, which is a fixed value. The light intensity value of the spectrum to be measured. To calibrate the light intensity values ​​of the dark noise spectrum, The light intensity value is the reference spectrum.

[0151] S15: Determine wafer thickness using the reflectivity of the wafer surface to detect the polishing endpoint based on wafer thickness.

[0152] In one embodiment, the reflectivity of the wafer surface can be compared with the theoretical reflectivity modeled based on the Fresnel formula to determine the wafer thickness at the current moment. The polishing endpoint can then be detected based on the wafer thickness at the current moment. If the polishing endpoint is reached, the polishing is complete. If the polishing endpoint is not reached, the polishing continues and the above-mentioned polishing endpoint detection steps are performed.

[0153] In summary, this application obtains the reflection spectrum of the wafer surface during chemical mechanical polishing (CMP) and corrects the reflection spectrum using a preset correction coefficient to obtain a corrected reflection spectrum. This correction coefficient is calibrated based on the thickness change of the window at different wear stages in the polishing environment corresponding to the CMP process. Based on the comparison results between the corrected reflection spectrum and the corresponding threshold, the corrected reflection spectrum is divided into a dark noise spectrum and a test spectrum. The wafer thickness is then determined based on the test spectrum, the dark noise spectrum, the reference spectrum, and the reflectance of the preset reference spectrum. The polishing endpoint is then detected based on this wafer thickness.

[0154] This application corrects the obtained reflectance spectrum using a preset correction coefficient to obtain a corrected reflectance spectrum. The endpoint detection process uses this corrected reflectance spectrum, which reduces deviations caused by measurement system errors, equipment errors, or environmental factors. This makes the reflectance spectrum used for endpoint detection closer to the theoretical value, thereby improving the accuracy of endpoint detection, avoiding over-polishing or under-polishing, and further improving the yield and polishing accuracy of wafer production.

[0155] The calibration process of the correction coefficient in this application is performed in a polishing environment corresponding to the chemical mechanical polishing process. This ensures that the spectrum used in the calibration process and the spectrum used for the end-point detection of the actual polishing process are the spectra under the corresponding environment. This effectively avoids the problem of inaccurate calibration of the correction coefficient due to environmental differences, improves the accuracy of the spectrum after correction based on the correction coefficient, thereby improving the accuracy and reliability of the end-point detection. It also avoids over-polishing or under-polishing, further improving the yield and polishing precision of wafer production.

[0156] The calibration process of the correction coefficient in this application takes into account the thickness change of the window at different wear stages. By utilizing the different thicknesses of the window at different polishing times, the intensity of the reflected spectrum obtained is different, and multiple sets of experimental data for calibration of the correction coefficient can be obtained, providing sufficient data support for the calibration of the correction coefficient, effectively ensuring the accuracy of the calibration of the correction coefficient, and further improving the accuracy of the spectrum after correction based on the correction coefficient, thereby improving the accuracy and reliability of the endpoint detection.

[0157] The effectiveness of the polishing endpoint detection method in this application can be achieved through... Figure 8 and Figure 9 The comparison shows that, among them Figure 8 This is a spectral fit curve of the polishing endpoint detection method using existing technology; Figure 9 The graph shows the spectral goodness-of-fit curve of the polishing endpoint detection method proposed in this application. The horizontal axis represents wavelength, and the vertical axis represents reflectance. The blue smooth curve (Model) represents the ideal spectral curve, and the red curve (Refl) represents the measured spectral curve. A comparison of the two curves shows that the measured spectral curve of the polishing endpoint detection method proposed in this application is closer to the ideal spectral curve. This application effectively improves the spectral goodness-of-fit, thereby improving the accuracy of endpoint detection.

[0158] See Figure 10 , Figure 10 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 10 The electronic device 300 shown in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304.

[0159] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0160] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0161] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0162] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation method described in the polishing endpoint detection method provided in the embodiments of this application, or they can execute the implementation method of the electronic device described in the embodiments of this application, which will not be repeated here.

[0163] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0164] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0165] This application provides a computer program product, which includes computer-executable instructions or a computer program. The computer-executable instructions or computer program are stored in a computer-readable storage medium. The processor of an electronic device reads the computer-executable instructions from the computer-readable storage medium and executes the computer-executable instructions, causing the electronic device to perform the polishing endpoint detection method described in this application embodiment.

[0166] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software 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 implementations should not be considered beyond the scope of this application.

[0167] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0168] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces or units, or they may be electrical, mechanical, or other forms of connection.

[0169] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0170] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0171] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for detecting the polishing endpoint, characterized in that, include: Obtain the reflection spectrum of the wafer surface during chemical mechanical polishing; The reflection spectrum is corrected using a preset correction factor to obtain the corrected reflection spectrum; The correction factor was calibrated based on the thickness variation of the window plate at different wear stages in the polishing environment corresponding to the chemical mechanical polishing process. The average light intensity, maximum light intensity, and spectral fit goodness of the corrected reflectance spectrum are determined; the corrected reflectance spectrum with the maximum light intensity greater than a first threshold or the average light intensity less than a second threshold, and the sampling point location is within a preset location range, is determined as the dark noise spectrum; The corrected reflectance spectrum that has an average light intensity or a maximum light intensity greater than a predetermined screening threshold and a spectral goodness of fit greater than a preset goodness of fit threshold is determined as the spectrum to be tested. The dark noise spectrum is removed from the spectrum to be measured to obtain the first spectrum; The dark noise spectrum is removed from the reference spectrum to obtain the second spectrum; The reflectance of the wafer surface is determined based on the first spectrum, the second spectrum, and the reflectance of a preset reference spectrum. The wafer thickness is determined using the reflectivity of the wafer surface, and the polishing endpoint is detected based on the wafer thickness.

2. The method as described in claim 1, characterized in that, Also includes: In the polishing environment corresponding to the chemical mechanical polishing process, the correction coefficient is calibrated based on the thickness change of the window at different wear stages, further including: Based on the polishing rate, polishing time, and initial thickness of the window, the theoretical thickness value of the window at different wear stages is determined; Based on the initial thickness of the window, the initial reflection spectrum, and the theoretical thickness value, the theoretical reflection spectrum of the window at different wear stages is determined; Obtain the actual reflectance spectra of the window at different wear stages; The correction coefficients are calibrated based on the principle of minimizing the error between the theoretical reflectance spectrum and the corresponding actual reflectance spectrum.

3. The method as described in claim 2, characterized in that, Also includes: The calibration period for the correction coefficient is determined based on the light intensity attenuation of the light source. The calibration operation is performed on the correction coefficient according to the calibration cycle.

4. The method as described in claim 3, characterized in that, Performing a calibration operation on the correction coefficient includes: The correction coefficient is calibrated based on the ratio of the initial light intensity to the current light intensity of the light source.

5. The method according to any one of claims 1 to 4, characterized in that, Also includes: In the polishing environment corresponding to the chemical mechanical polishing process, a reference spectrum is determined, including: In a polishing environment corresponding to the chemical mechanical polishing process, a standard wafer corresponding to the wafer is subjected to chemical mechanical polishing to obtain a reference reflectance spectrum of the surface of the standard wafer at different times during the polishing process; The reference reflectance spectrum is corrected using a preset correction factor to obtain the corrected reference reflectance spectrum; Determine the reference average light intensity and reference maximum light intensity of the corrected reference reflectance spectrum; The reference average light intensity and the reference maximum light intensity are compared with a predetermined screening threshold, and the corrected reference reflectance spectrum whose comparison result is greater than the screening threshold is used as the screening spectrum. The reference spectrum is determined based on the screened spectrum.

6. The method as described in claim 1, characterized in that, Also includes: Determine the average light intensity of all dark noise spectra to obtain the calibrated dark noise spectrum; The dark noise spectrum is removed from the spectrum to be measured to obtain a first spectrum, and the dark noise spectrum is removed from the reference spectrum to obtain a second spectrum, including: The calibration dark noise spectrum is removed from the spectrum to be measured to obtain the first spectrum; The calibration dark noise spectrum is removed from the reference spectrum to obtain the second spectrum.

7. A chemical mechanical polishing apparatus, characterized in that, include: Polishing platform, polishing pad, polishing head, and endpoint detection system; the endpoint detection system includes a light source, a spectrum receiver, and a controller. The polishing head is used to load the wafer and abut it against the polishing pad above the polishing platform; The light source is used to output optical signals to the wafer surface; The spectral receiver is used to receive the spectrum reflected from the wafer surface; The controller is used to perform the method as described in any one of claims 1 to 6 during the chemical mechanical polishing process of the wafer.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 6.

Images