Quality factor measuring and calculating method, wafer thickness measuring method and wafer thickness measuring device

Abstract

The invention provides a quality factor measuring and calculating method and a wafer thickness measuring method and device.A quality factor is used for evaluating the thickness value, determined by detecting an interference spectrum, of a wafer, and the quality factor measuring and calculating method comprises the steps that the interference spectrum is obtained; determining the thickness value of the wafer according to the interference spectrum; calculating a first quality factor corresponding to the thickness value of the wafer according to a power spectrum corresponding to the interference spectrum; determining a compensation factor according to the thickness value of the wafer; and correcting the first quality factor according to the compensation factor to obtain a quality factor, so that the quality factor is not influenced by the thickness value difference of the wafer. The effectiveness evaluation of the measured and calculated quality factor on the thickness measurement result of the wafer is more accurate.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of semiconductor chip processing. Specifically, it relates to a quality factor calculation method, a wafer thickness measurement method and device. BACKGROUND

[0002] In semiconductor chip processing, a wafer is processed by using a thinning process (Grinding) or a chemical mechanical polishing (CMP) process to remove a part of the wafer surface to make the wafer meet the surface processing requirements of the next process. The thickness and uniformity of the wafer directly affect the performance of the semiconductor device. Therefore, during the wafer processing, accurately measuring the thickness of the wafer in situ is crucial to improve the yield of chip manufacturing.

[0003] At present, the optical in-situ measurement technology of the thickness of the semiconductor silicon wafer generally relies on interference spectrum, and the optical path difference principle is used to measure the thickness of the silicon wafer. However, in the actual measurement process, due to the poor measurement environment or the attenuation of the light source and other factors, invalid measurement results may occur in the data of the interference spectrum. In order to evaluate the effectiveness of the measurement results, a quality factor is introduced to represent the confidence of the thickness measurement.

[0004] However, the resolution of the sensor, the thickness of the wafer, and the measurement noise will all affect the quality factor, resulting in that the quality factor cannot accurately represent the confidence of the thickness measurement, and the quality factors of wafers of different thicknesses cannot be directly compared, so that the effectiveness of the measurement results cannot be accurately evaluated, and the performance of the semiconductor device is affected. SUMMARY

[0005] Therefore, the present application provides a quality factor calculation method, a wafer thickness measurement method and device, and a wafer processing apparatus, so as to solve or at least alleviate one or more of the above problems and other aspects in the prior art.

[0006] The first aspect of the present application provides a quality factor calculation method, the quality factor is used to evaluate the thickness value of the wafer determined by detecting the interference spectrum, comprising:

[0007] Obtaining an interference spectrum;

[0008] Determining the thickness value of the wafer according to the interference spectrum;

[0009] Calculating a first quality factor corresponding to the thickness value of the wafer according to the power spectrum corresponding to the interference spectrum;

[0010] Determining a compensation factor according to the thickness value of the wafer;

[0011] The first quality factor is corrected according to the compensation factor to obtain a quality factor, so that the quality factor is not affected by the thickness value difference of the wafer.

[0012] Optionally, the obtaining the interference spectrum comprises:

[0013] Obtaining a first interference spectrum corresponding to wavelength and light intensity;

[0014] Filtering the first interference spectrum to remove the direct current component;

[0015] Obtaining a second interference spectrum corresponding to wave number and light intensity, the wave number being the reciprocal of the wavelength;

[0016] Interpolating the second interference spectrum to obtain the interference spectrum.

[0017] Optionally, the calculating the first quality factor corresponding to the thickness value of the wafer according to the power spectrum corresponding to the interference spectrum comprises:

[0018] Determining the power spectrum according to the interference spectrum;

[0019] Determining the corresponding power spectrum value according to the thickness value of the wafer and the power spectrum;

[0020] Normalizing the power spectrum value, and taking the normalization result as the first quality factor.

[0021] Optionally, the determining the compensation factor according to the thickness value of the wafer comprises:

[0022] Determining the compensation factor according to the thickness value of the wafer and a compensation function, the compensation function being a function with the thickness of the wafer as the independent variable, and the compensation factor being fitted according to the thickness value of the wafer and the first quality factor of the wafer with different thicknesses by the following steps:

[0023] Selecting a plurality of wafers with different thicknesses, wherein the thickness value of the wafer is known;

[0024] Measuring the first quality factor of each wafer under the same measurement state;

[0025] Polynomial fitting the thickness value of the plurality of wafers and the corresponding first quality factor to obtain a thickness-quality factor curve, and the function corresponding to the curve being the compensation function;

[0026] Determining the reciprocal of the compensation function value corresponding to the thickness value of the wafer as the compensation factor.

[0027] Optionally, the correcting the first quality factor according to the compensation factor to obtain a quality factor comprises:

[0028] Correcting the first quality factor according to the compensation factor to obtain a second quality factor;

[0029] calculating a prediction interval of the thickness value of the wafer according to an initial thickness of the wafer, a material removal rate and a processing time;

[0030] judging whether the thickness value of the wafer falls into the prediction interval;

[0031] if the thickness value of the wafer falls into the prediction interval, amplifying the second quality factor to be the quality factor;

[0032] if the thickness value of the wafer does not fall into the prediction interval, taking the second quality factor as the quality factor.

[0033] Optionally, the method further comprises:

[0034] repeating the calculation to obtain a plurality of quality factors;

[0035] taking an average value of the plurality of quality factors as the quality factor.

[0036] A second aspect of the present application provides a method for measuring the thickness of a wafer, comprising:

[0037] measuring the thickness value of the wafer according to the quality factor calculation method of the first aspect and calculating the corresponding quality factor;

[0038] taking the thickness value of the wafer with the corresponding quality factor not less than a preset threshold as the measurement result.

[0039] A third aspect of the present application provides a device for measuring the thickness of a wafer, configured to execute the method for measuring the thickness of a wafer of the first aspect, comprising:

[0040] a light source configured to generate a measurement spectrum;

[0041] a probe configured to emit the measurement light and receive the reflected light;

[0042] a sensor configured to receive the reflected light and generate the interference spectrum;

[0043] a control unit configured to control the light source to generate the measurement spectrum and control the sensor to receive the interference spectrum.

[0044] A fourth aspect of the present application provides a wafer processing apparatus, comprising a wafer processing device, a controller and a device for measuring the thickness of a wafer according to the third aspect;

[0045] the controller controls the wafer processing device to process the surface of the wafer while controlling the device for measuring the thickness of a wafer to execute the method for measuring the thickness of a wafer according to the second aspect to obtain a measurement result according to the interference spectrum, and adjusts the processing recipe of the wafer processing device to process the surface of the wafer according to the measurement result.

[0046] Optionally, the wafer processing device is a chemical mechanical polishing device or a thinning device.

[0047] The quality factor measurement method provided by the application has the following technical effects: the quality factor measured by the quality factor measurement method is significantly reduced in influence of system hardware limitation, measurement error and noise, so that the effectiveness evaluation of the wafer thickness measurement result is more accurate, and the quality factors of wafer samples with different thicknesses can be uniformly compared. The wafer thickness measurement method and device provided by the application can measure the thickness value of the wafer after quality factor screening, which is closer to the actual physical result, so as to be used as the basis for adjusting the processing formula of the wafer surface in the wafer processing device, so as to improve the wafer surface processing precision and improve the chip manufacturing yield. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0049] Figure 1 is a schematic diagram of an embodiment of the quality factor measurement method of the application.

[0050] Figure 2 is a schematic diagram of the calibration result of the first quality factor of wafers with different thicknesses.

[0051] Figure 3 is Figure 2 is a schematic diagram of the second fitting of the thickness-quality factor curve of the calibration result.

[0052] Figure 4 is a schematic diagram of another embodiment of the quality factor measurement method of the application.

[0053] Figure 5 is a schematic diagram of an embodiment of the wafer thickness measurement method of the application.

[0054] Figure 6 is a schematic diagram of the measurement result without quality factor screening in wafer thinning.

[0055] Figure 7 is a schematic diagram of the measurement result with quality factor screening. Figure 6

[0056] Figure 8 is a schematic diagram of the wafer thickness measurement device of the application.

[0057] Figure 9 is a schematic diagram of the wafer processing device of the application. DETAILED DESCRIPTION ​

[0058] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art should fall within the protection scope of the present invention.

[0059] In the description of this invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0060] In addition, in the description of this invention, unless otherwise specified and limited, it should be noted that the terms "installation", "connection" and "linking" should be interpreted broadly. For example, they can refer to mechanical or electrical connections, or internal connections between two components. They can be direct connections or indirect connections through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0061] This invention provides a method for calculating the quality factor in optical in-situ measurement of wafer thickness, wherein the quality factor is used to evaluate the wafer thickness determined by detecting interference spectra.

[0062] It should be noted that optical in-situ measurement technology can only be used to measure the thickness of non-metallic films on wafers. In this invention, the thickness of a wafer can refer to the thickness of a bare wafer or the thickness of the non-metallic film on the wafer after thin film deposition, especially the real-time thickness of the non-metallic film during the removal of the non-metallic film.

[0063] like Figure 1 As shown, the quality factor calculation method provided in this embodiment of the invention includes the following steps:

[0064] S1. Obtain the interference spectrum I.

[0065] After the light source projects light onto the wafer surface, part of the light is directly reflected from the upper surface of the wafer, while the other part is refracted by the upper surface and reflected from the lower surface. The reflected light from the upper and lower surfaces of the wafer can be received by a sensor. The two parts of reflected light interfere due to the optical path difference, thus obtaining the interference spectrum I.

[0066] S2. Determine a thickness value d0 of the wafer according to the interference spectrum.

[0067] The oscillation frequency and amplitude of the interference spectrum are determined by the thickness d of the wafer, the refractive index n of the wafer and the wavelength λ, so the thickness of the wafer can be calculated by converting the spectrum signal to the frequency domain through Fourier transform to extract the characteristic peak frequency in the frequency spectrum F(ω), or the thickness of the wafer can be determined according to the scheme disclosed in Chinese patent application CN118875959A.

[0068] However, the resolution of the sensor directly affects the accuracy of the Fourier transform result, and lower resolution leads to loss or ambiguity of spectral information, affecting the calculation accuracy of the film thickness; and the Fourier transform is difficult to distinguish the characteristic frequency of the measured noise and the true thickness, resulting in obvious errors in the measurement of the film thickness. If the film thickness data with obvious errors is used as the basis for processing the wafer surface, the thickness and uniformity of the wafer will not meet the surface processing requirements of the next process, and will directly lead to a decrease in the yield of the chip.

[0069] S3. Calculate a first quality factor Q1 corresponding to the thickness value of the wafer according to the power spectrum corresponding to the interference spectrum.

[0070] The frequency spectrum obtained by Fourier transform of the spectrum signal contains amplitude and phase information, and the characteristic peak in the frequency spectrum may represent the thickness of the wafer or noise. The corresponding power spectrum only retains the amplitude information, so that the first quality factor representing the confidence of the thickness can be calculated to evaluate the effectiveness of the measurement result and to filter out the film thickness data with obvious errors.

[0071] However, the first quality factor also comes from the measured spectrum, and the resolution of the sensor, the thickness of the wafer and the measurement noise will also affect the accuracy of the quality factor, resulting in errors when characterizing the confidence of the thickness measurement. Moreover, the first quality factors calculated for wafers of different thicknesses differ greatly, even by orders of magnitude, making them unable to be directly compared with each other, and thus a unified threshold cannot be established to evaluate the effectiveness of the measurement result.

[0072] S4. Determine a compensation factor δ according to the thickness value d0 of the wafer.

[0073] Even under the best measurement conditions, as the thickness increases, the quality factor decreases, that is, the resolution of the sensor has a greater impact on the quality factor of thicker wafers. In order to balance the impact of the resolution of the sensor and the thickness of the wafer on the quality factor, a compensation factor is determined according to the thickness value of the wafer to correct the quality factor.

[0074] S5. Correct the first quality factor Q1 according to the compensation factor δ to obtain a quality factor Q, so that the quality factor is not affected by the thickness difference of the wafer.

[0075] It is easy to understand that the first quality factor Q1 correction mode depends on the determination mode of the compensation factor δ, for example, when the compensation factor δ represents a relative difference, the correction can be made by Q = Q1 + δ; when the compensation factor δ represents a relative ratio, the correction can be made by Q = CQ1; or other correction modes can be used to correct the first quality factor Q1 to obtain the quality factor Q to compensate the influence of the wafer thickness change in the wafer processing process on the quality factor. Different thickness values of the wafer correspond to different values of the compensation factor, so that the first quality factor corresponding to the thickness of different wafers is corrected to obtain the quality factor, so that the quality factor is not affected by the thickness difference of the wafer, and the quality factors of wafers with different thicknesses can be kept in the same order of magnitude.

[0076] Optionally, step S1 further comprises the following steps:

[0077] S11. Obtain a first interference spectrum I(λ) corresponding to the wavelength and light intensity, I(λ) is a function of the frequency signal composition of the first interference spectrum, and λ is the wavelength.

[0078] S12. Filter out the direct current component of the first interference spectrum I(λ).

[0079] The direct current component of the interference spectrum corresponds to the average value of the light intensity (i.e. the zero frequency superposition of all frequency components), which only represents the total light intensity background and does not contain any spectrum information related to the wavelength or frequency (such as molecular absorption, scattering characteristics, etc.). The actual spectrum information is carried by the alternating current component (non-zero frequency part) of the interference signal, and filtering out the direct current component can avoid invalid data occupying computing resources.

[0080] Specifically, the first interference spectrum I(λ) is frequency-division processed by using a frequency-division algorithm (for example, an empirical mode decomposition algorithm), to obtain a direct current component f D (λ), a non-0 frequency component f i (λ), and a residual frequency component f res (λ). The direct current component f D (λ) is filtered out, and according to the Fresnel law, the first interference spectrum function I(λ) after filtering out the direct current component can be expressed as A is a constant coefficient, n is the refractive index of the wafer, and d is the thickness of the wafer.

[0081] S13. Obtain a second interference spectrum

[0082] The wave number is the reciprocal of the wavelength, and in the wavelength domain, the resolution of the spectrum changes with the wavelength, while the wave number domain resolution is consistent, so that the analysis process can be simplified.

[0083] S14. Interpolating the second interference spectrum, obtaining the interference spectrum.

[0084] In the wave number domain, since the data is discrete, in order to obtain a more accurate function representation, the discrete second interference spectrum is interpolated, thereby generating a continuous function as the interference spectrum, so that the signal value at any wave number can be estimated. Preferably, spline interpolation can be performed by the maximum wavelength and the minimum wavelength that the sensor can collect.

[0085] Optionally, steps S2 and S3 further comprise the following steps:

[0086] The second interference spectrum in the wave number domain Performing fast Fourier transform (FFT) to obtain the interference spectrum F(ω);

[0087] Using a peak searching function to extract the peak position ω0 in the interference spectrum, and inputting the calculation formula of the thickness of the wafer Obtaining the thickness value d0 of the wafer

[0088] According to the interference spectrum F(ω), the power spectrum P(ω) is determined, and the power spectrum P(ω) is the square of the interference spectrum F(ω), that is

[0089] P(ω)=F(ω) 2 ;

[0090] The power spectrum value corresponding to the thickness value d0 of the wafer is P(ω0)=F(ω0) 2 , and the normalized algorithm is used for normalization processing, and the result is taken as the first quality factor Q1(d0). The formula of the normalization algorithm is

[0091] The first quality factor after normalization processing falls within the range of [0, 1], which is conducive to the direct comparison of the quality factors of wafers with different thicknesses.

[0092] Optionally, step S4 further comprises the following steps:

[0093] S41. Determining the compensation factor δ according to the thickness value d0 of the wafer and the compensation function f(d), specifically inputting the thickness value d0 of the wafer into the formula δ=g(f(d)) for calculating the compensation factor, and the compensation function f(d) is a function with the thickness d of the wafer as the independent variable.

[0094] In order to balance the influence of the resolution of the sensor and the thickness of the wafer on the quality factor, the compensation function f(d) should be able to make appropriate compensation for the first quality factor corresponding to different thickness values of the wafer, so the first quality factor of the wafer with different thickness can be calibrated, and then the compensation function f(d) is obtained by function fitting, so that the compensation factor δ=g(f(d0)) can be determined at any thickness value d0 of the wafer, and the specific steps are as follows:

[0095] S411. Select a plurality of wafers with different thicknesses, the thickness values of which can be measured in advance by other means, and the thickness value range can cover the thickness variation range of the wafer in the processing process as much as possible.

[0096] S412. Measure the first quality factor of each wafer under the same measurement state, preferably under the best measurement condition in the laboratory, that is, adjust the measurement probe to be perpendicular to the wafer, and make the upper surface of the wafer located at the focal point of the measurement probe, measure the interference spectrum of each wafer, and then convert the power spectrum to obtain the first quality factor of each wafer.

[0097] S413. Fit the thickness values and the first quality factors of the plurality of wafers with different thicknesses in a polynomial fitting manner to obtain a thickness-quality factor curve, and the corresponding function f(d) is the compensation function. It is found through experiments that with the increase of the thickness of the wafer, the calculated first quality factor shows a trend of first increasing and then decreasing, so the compensation function f(d) generated by the quadratic fitting is preferably selected, so as to be more objective in accordance with the change law of the first quality factor, and the compensation result is more accurate.

[0098] S414. Determine the reciprocal of the compensation function value f(d0) corresponding to the thickness value d0 of the wafer as the compensation factor, that is,

[0099] In an ideal state, the theoretical value of the first quality factor of the wafer with a known thickness corresponding to its thickness value is 1, and the compensation factor at this time represents the relative comparison deviation of the calculated first quality factor and the theoretical value. Therefore, in step S5, the quality factor

[0100] Figure 2 For the first quality factor calibration results of 7 wafers with an optical thickness of 30-2800 μm under the best measurement condition in the laboratory, the thickness-quality factor curve obtained by quadratic fitting of the calibration results is shown in Figure 3 The compensation function f(d) obtained by fitting is

[0101] It should be noted that, ideally, the quality factor Q reflects the energy ratio of the main peak of the signal and the noise, and should be determined by the instrument noise condition and the material quality, and is irrelevant to the thickness. However, in practice, it is found that the quality factors of wafer samples with different thicknesses have large differences, and even are not in the same order of magnitude. Through repeated experiments and verification, it is determined that the non-ideal factors such as sensor resolution, detector response curve, and system energy broadening will make the interference signal of the thick sample wider or the peak value distorted, resulting in a low quality factor Q.

[0102] The first quality factor calculated by the wafer with different thicknesses is compensated according to the thickness in the embodiment, so that the quality factor is not affected by the thickness difference of the wafer, that is, the measurement error introduced by the system hardware limitation is corrected, the calculated quality factor Q is closer to the physical definition, and the effectiveness evaluation of the thickness measurement result of the wafer is more accurate. Also, the quality factors Q of wafer samples with different thicknesses can be compared uniformly.

[0103] As shown in Figure 4 Another embodiment of the present application provides a quality factor calculation method, which further corrects the quality factor obtained in step S5 on the basis of the previous embodiment. Specifically, the following steps are included:

[0104] S5’. The first quality factor Q1 is corrected according to the compensation factor δ to obtain the second quality factor Q2, and the correction method is the same as step S5.

[0105] S6. The initial thickness of the wafer, the material removal rate, and the wafer processing time are used to calculate the predicted interval of the thickness value d0 of the wafer.

[0106] The initial thickness d ini of the wafer and the wafer processing time t are known, and the material removal rate v is fixed for a specific wafer material and processing formula, which can be determined by pre-processing the same wafer, so as to infer the thickness value d'0=d ini -vt of the wafer after a predetermined processing time t. That is, the starting time of wafer processing is taken as 0 time, and the thickness value d0 of the wafer determined by steps S1 and S2 at t0 time should be theoretically consistent with the thickness value d'0 of the wafer inferred according to the processing time, that is, d0=d'0.

[0107] Due to the influence of material removal rate measurement and wafer processing process control accuracy, the thickness of the wafer inferred according to the processing time is not accurate, and the conversion into a predicted interval can more accurately predict the range of the wafer after processing. The predicted interval is represented as or B=[α*d'0,β*d'0], where α and β are constant values determined according to the material removal rate v, system error, etc., for example, α=0.9 and β=1.1. or a = 0.9 and b = 1.1, etc.

[0108] S7. Determine whether the thickness value of the wafer falls into the predicted interval, that is, determine whether the expression d0 e B is true. If true, it means that the thickness value of the wafer falls into the predicted interval, indicating that the confidence of the measurement result is higher, and then the second quality factor Q2 is enlarged by a certain proportion to serve as the quality factor Q, that is, Q = k * Q2, k is a constant greater than 1; if not true, it means that the thickness of the wafer does not fall into the predicted interval, indicating that the confidence of the measurement result is low, and then the second quality factor Q2 is taken as the quality factor Q, that is, Q = Q2.

[0109] In this embodiment, by determining whether the thickness value of the wafer falls into the predicted interval, the confidence of the measurement result is evaluated from another dimension, and then the quality factor corresponding to the measurement result falling into the predicted interval is enlarged, so as to further improve the compensation ability of the quality factor for the quality factor caused by system error, measurement noise, etc. and ensure that the effectiveness evaluation of the wafer thickness measurement result by the quality factor is more accurate, thereby improving the chip manufacturing yield.

[0110] In the quality factor measurement method provided in another embodiment, steps S4 and S5 are not performed, that is, there is no need to determine the compensation factor and correct the first quality factor, but the second quality factor is directly equal to the first quality factor, that is, Q2 = Q1, and then steps S6 and S7 are performed to determine the quality factor Q.

[0111] This embodiment can also compensate for the quality factor caused by system error, measurement noise, etc. and make the effectiveness evaluation of the wafer thickness measurement result by the quality factor more accurate, thereby improving the chip manufacturing yield.

[0112] Optionally, in the quality factor measurement method provided in any one of the embodiments, the following steps are further included:

[0113] repeatedly measuring to obtain multiple quality factors;

[0114] taking the average value of the multiple quality factors as the quality factor.

[0115] In actual measurement, the quality factor will be affected by external random noise, interference signal, device transient instability, etc., so as to appear abnormal values. These abnormal values are often accidental and do not appear continuously, and the deviation is usually not extremely large. After multiple repeated measurements, the average value is taken, so as to make the quality factor of random error close to a lower value, and the quality factor of effective signal is stable at a higher value, thereby helping to reduce the influence of random error and making the result closer to the true value (in line with the law of large numbers).

[0116] The application also provides a wafer thickness measurement method, in one embodiment, as shown in Figure 5 the following steps are included:

[0117] S101. The thickness value d0 of the wafer is measured according to the quality factor calculation method of the application, and the corresponding quality factor Q is calculated.

[0118] S102. The measured thickness value of the wafer is screened according to the quality factor Q, and the thickness value of the wafer corresponding to the quality factor not less than the preset threshold value is taken as the measurement result.

[0119] The thickness and quality factor of the wafer are two important parameters when the wafer is thinned or polished. Taking the thickness measurement of the wafer thinned from 35 μm to 5 μm as an example, Figure 6 the measurement result without quality factor screening is shown, and there are multiple obvious abnormal values. If this is used as the basis for judging the wafer processing process, for example, the wafer reaches the predetermined thickness as the condition for ending the wafer processing in the end point detection, it will undoubtedly lead to uneven quality of the processed wafers, affecting the chip manufacturing yield.

[0120] As a comparison, Figure 7 the measurement result after quality factor screening is shown, and the preset threshold value is 0.3, that is, only the thickness value of the wafer corresponding to the quality factor greater than or equal to 0.3 is taken as the measurement result. It can be seen that after the quality factor screening, most of the abnormal values are filtered out, especially the abnormal values with large errors are completely filtered out. The filtered thickness measurement result is more consistent with the actual change trend of the thickness of the wafer, effectively improving the reliability of the measurement, and the thickness and uniformity of the wafer processed by the measurement result are guaranteed, improving the chip manufacturing yield.

[0121] The application also provides a wafer thickness measurement device for executing the wafer thickness measurement method of the application.

[0122] As shown in Figure 8 the thickness measurement device 10 provided by the embodiment of the application includes:

[0123] a light source 1 for generating a measurement spectrum, preferably a wide spectrum light source;

[0124] a probe 2 for emitting measurement light to the measured wafer and receiving reflected light of the measured wafer, the measurement light and the reflected light being emitted and received through the objective lens of the probe 2;

[0125] a sensor 3 for receiving the reflected light and generating an interference spectrum, the sensor 3 being preferably a spectrometer;

[0126] a control unit 4 for controlling the light source to generate the measurement spectrum and controlling the sensor to receive the interference spectrum.

[0127] The transmission path (i.e. optical path) of the light in the thickness measurement device 10 is shown by arrows, at least part of the transmission path is an optical fiber, and the transmission path can further include a beam splitter 5.

[0128] The present application also provides a wafer processing device for processing a wafer surface.

[0129] As shown in Figure 9 The wafer processing device 100 provided by the embodiments of the present application comprises a wafer processing device 20, a controller 30 and a thickness measurement device 10.

[0130] The controller 30 controls the wafer processing device 20 to process the wafer surface while controlling the thickness measurement device 10 to perform the wafer thickness measurement method of the present application to obtain a measurement result, and adjusts the recipe of the wafer processing device 20 to process the wafer surface according to the measurement result.

[0131] Optionally, the wafer processing device 20 is a chemical mechanical polishing device or a thinning device. The chemical mechanical polishing device uses a carrier head to press the wafer onto the surface of a polishing pad, and relies on the relative movement between the wafer and the polishing pad and the abrasive particles in the polishing liquid to achieve polishing of the wafer surface. The thinning device grinds the wafer surface by relative rotation of the grinding wheel to process the wafer to a predetermined thickness. Both of them involve reducing the impact of wafer thickness variation and surface topography while removing the material layer on the wafer surface.

[0132] The wafer processing device 100 of the present application is based on the measurement result of the thickness measurement device 10 and the thickness measurement method of the present application, which ensures that the thickness value of the wafer measured after quality factor screening is closer to the actual physical result, thereby obtaining higher quality wafer surface processing precision and improving chip manufacturing yield.

[0133] The above embodiments are only used to illustrate the embodiments of the present application, but not to limit the embodiments of the present application. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the embodiments of the present application. Therefore, all equivalent technical solutions belong to the scope of the embodiments of the present application, and the patent protection scope of the embodiments of the present application should be defined by the claims.

Claims

1. A method of measuring a quality factor, characterized by, A quality factor is used to evaluate a thickness value of a wafer determined by detecting an interference spectrum, comprising: obtaining an interference spectrum; determining a thickness value of the wafer according to the interference spectrum; calculating a first quality factor corresponding to the thickness value of the wafer according to a power spectrum corresponding to the interference spectrum; determining a compensation factor according to the thickness value of the wafer; correcting the first quality factor according to the compensation factor to obtain the quality factor, so that the quality factor is not affected by the difference of the thickness value of the wafer.

2. The quality factor measurement method of claim 1, wherein, The obtaining of the interference spectrum comprises: obtaining a first interference spectrum corresponding to wavelength and light intensity; filtering out the direct current component of the first interference spectrum; obtaining a second interference spectrum corresponding to wave number and light intensity, the wave number being the reciprocal of the wavelength; interpolating the second interference spectrum to obtain the interference spectrum.

3. The quality factor measurement method of claim 2, wherein, The calculating of the first quality factor corresponding to the thickness value of the wafer according to the power spectrum corresponding to the interference spectrum comprises: determining the power spectrum according to the interference spectrum; determining a power spectrum value corresponding to the thickness value of the wafer and the power spectrum; normalizing the power spectrum value, and taking the normalization result as the first quality factor.

4. The method of claim 3, wherein the quality factor is calculated by: ###00002### Q = 2πfL R The determining of the compensation factor according to the thickness value of the wafer comprises: determining the compensation factor according to the thickness value of the wafer and a compensation function, the compensation function being a function with the thickness of the wafer as the independent variable, and the compensation function being fitted according to the thickness value of the wafer and the first quality factor of the wafer with different thicknesses by the following steps: selecting a plurality of wafers with different thicknesses, wherein the thickness value of the wafer is known; measuring the first quality factor of each wafer under the same measurement state; polynomial fitting the thickness value and the corresponding first quality factor of the plurality of wafers to obtain a thickness-quality factor curve, and the function corresponding to the curve being the compensation function; determining the reciprocal of the compensation function value corresponding to the thickness value of the wafer as the compensation factor.

5. The quality factor measurement method according to any one of claims 1 to 4, characterized in that, The correcting of the first quality factor according to the compensation factor to obtain the quality factor comprises: correcting the first quality factor according to the compensation factor to obtain a second quality factor; calculating a prediction interval of the thickness value of the wafer according to the initial thickness of the wafer, the material removal rate and the processing time; judging whether the thickness value of the wafer falls into the prediction interval; if the thickness value of the wafer falls into the prediction interval, then the second quality factor is enlarged and taken as the quality factor; if the thickness value of the wafer does not fall into the prediction interval, then the second quality factor is taken as the quality factor.

6. The method of claim 5, wherein the quality factor is calculated by: ###0001### where: Q = quality factor; f = frequency; R = resistance; L = inductance; and C = capacitance. Further comprising: repeating the measurement to obtain a plurality of quality factors; taking the average value of the plurality of quality factors as the quality factor.

7. A method of measuring the thickness of a wafer, characterized by, Comprising: measuring the thickness value of the wafer according to the quality factor measurement method of any one of claims 1-6, and measuring the corresponding quality factor; taking the thickness value of the wafer with the corresponding quality factor not less than a preset threshold as the measurement result.

8. A wafer thickness measuring device, characterized by comprising: For performing the wafer thickness measurement method of claim 7, comprising: a light source for generating a measurement spectrum; a probe for emitting measurement light and receiving reflected light; a sensor for receiving reflected light and generating the interference spectrum; a control unit for controlling the light source to generate the measurement spectrum and controlling the sensor to receive the interference spectrum.

9. A wafer processing apparatus, characterized by comprising: Comprising: a wafer processing device, a controller and the wafer thickness measurement device of claim 8; The controller controls the wafer processing device to process the surface of the wafer, controls the thickness measuring device of the wafer to perform the thickness measuring method of the wafer as claimed in claim 7 to obtain a measuring result, and adjusts the processing recipe of the wafer processing device to process the surface of the wafer according to the measuring result.

10. The wafer processing apparatus of claim 9, wherein, The wafer processing device is a chemical mechanical polishing device or a thinning device.

Images