CMP apparatus and polishing method

Abstract

To provide a CMP apparatus and a polishing method that can easily improve the accuracy of measuring the thickness of a polishing layer among a plurality of layers stacked on the surface of a wafer.SOLUTION: There is provided a CMP apparatus 1 for polishing a polishing layer W1 of a wafer W comprising the polishing layer W1 and at least one underlayer W2 located below the polishing layer. The CMP apparatus comprises: a database 30 including respective model waveforms of reflected light spectra for respective thicknesses of the polishing layer corresponding to respective thicknesses of the underlayer; a detection unit 120 for determining the thickness of the polishing layer on the basis of the database and a first reflected light spectrum detected during polishing of the wafer; and an arithmetic processing unit 130 for generating the database on the basis of information on materials of the polishing layer, information on materials of the underlayer, a first thickness range of the polishing layer, a second thickness range of the underlayer, and thickness intervals of the polishing layer and the underlayer.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a CMP apparatus that polishes a polishing layer (or measurement layer) having a polishing surface formed on a wafer surface, such as an oxide film, by chemical mechanical polishing (CMP), and more particularly to a CMP apparatus and polishing method that determine film thickness based on optical information contained in light reflected from the wafer. [Background technology]

[0002] A CMP apparatus polishes a polishing layer (or measurement layer) having a polishing surface formed on the surface of a wafer, such as an oxide film, etc. The CMP apparatus polishes the wafer by rotating a polishing pad attached to a platen, which is a polishing surface plate, and a wafer held in a polishing head, and pressing the wafer against the polishing pad while supplying slurry between them.

[0003] Patent Document 1 discloses that, as a means of observing the thickness of the polishing layer (or measurement layer) formed on the wafer surface in-situ during polishing, light is projected onto the wafer being polished, the reflected light is received, and the light is compared with a pre-stored normal pattern.

[0004] Furthermore, Patent Document 2 discloses that a reflected light spectrum from the wafer surface is generated, and in order to reduce the effect of variations in the reflected light spectrum, three-dimensional data consisting of multiple spectra arranged along the polishing time is created, and the film thickness of the substrate is determined based on the three-dimensional data.

[0005] Furthermore, Patent Document 2 describes that the reflected light spectrum may also be affected by the underlying structure (underlying layer) present below the film (exposed film) that constitutes the exposed surface of the substrate, and describes that a plurality of reference data measured using a plurality of reference substrates with different underlying structures is provided, and the position of the data area within the reference data that most closely matches the three-dimensional data created during polishing of the substrate is determined, and the film thickness associated with this determined position is then determined. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2020-188233 [Patent Document 2] Patent Publication No. 2021-91038 Summary of the Invention [Problem to be solved by the invention]

[0007] In recent years, semiconductors have become increasingly miniaturized and layered, resulting in more stringent requirements for film thickness measurement. When measuring the thickness of a polishing layer (or measurement layer), which is the uppermost layer among multiple films layered on a wafer surface, there is a risk that the thickness of the base layer formed below the polishing layer (or measurement layer) may affect the measured thickness of the polishing film. Furthermore, as described in Patent Document 2, when multiple reference data are generated using multiple reference substrates with different underlying structures, if changes occur in the measurement conditions or specifications, it becomes necessary to accumulate reference data using additional reference substrates, which presents a problem of difficulty in responding to such changes.

[0008] An object of the present invention is to provide a CMP apparatus and a polishing method that can easily improve the accuracy of measuring the thickness of a polishing layer among multiple layers stacked on the surface of a wafer. [Means for solving the problem]

[0009] The first CMP apparatus of the present invention is a CMP apparatus for polishing a polishing layer of a wafer having a polishing layer and at least one underlayer located below the polishing layer, comprising: a database containing model waveforms of reflected light spectra for each thickness of the polishing layer corresponding to each thickness of the underlayer; a detection unit that determines the thickness of the polishing layer based on the database and a first reflected light spectrum detected during polishing of the wafer; and an arithmetic processing unit that generates the database based on information on the material of the polishing layer, information on the material of the underlayer, a first thickness range of the polishing layer, a second thickness range of the underlayer, and the thickness intervals of the polishing layer and the underlayer.

[0010] A second CMP apparatus of the present invention is the first CMP apparatus, wherein the detection unit stops polishing of the wafer when the first reflected light spectrum matches or is similar to a model waveform in the database.

[0011] A third CMP apparatus of the present invention is the first CMP apparatus, wherein the detection unit stops polishing the wafer when the first reflected light spectrum matches or resembles a model waveform in the database that corresponds to a thickness equal to or less than a target thickness.

[0012] A fourth CMP apparatus of the present invention is a CMP apparatus according to the second or third CMP apparatus, wherein the first thickness range includes an initial film thickness value of the polishing layer at the start of the polishing operation and a target film thickness value of the polishing layer after the polishing operation is completed, and the second thickness range includes an error range that occurs when manufacturing the base layer.

[0013] The polishing method of the present invention is a polishing method applied to a CMP apparatus for polishing a polishing layer of a wafer having a polishing layer and at least one underlayer located below the polishing layer, the method comprising: This is a polishing method that generates a database based on information about the material of the polishing layer, information about the material of the base layer, a first thickness range of the polishing layer, a second thickness range of the base layer, and thickness intervals between the polishing layer and the base layer, and determines the thickness of the polishing layer based on the database and a first reflected light spectrum detected during polishing of the wafer. [Effects of the Invention]

[0014] According to the present invention, model waveforms of the reflected light spectrum of the surface layer for each thickness, which are combinations of the thickness range of the surface layer and the thickness range of the underlayer, are stored as a database, the database is referenced to select from the model waveforms the model waveform that is most similar to the reflected light spectrum acquired during polishing, the thickness of the surface layer is determined from the selected model waveform, and polishing is stopped when the determined thickness of the surface layer becomes equal to or less than the target thickness.Therefore, in a wafer with a layered structure on which multiple films are formed, the thickness of the surface layer can be accurately controlled in situ without being affected by errors in the thickness value of the underlayer. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a perspective view of a main part of a CMP apparatus according to an embodiment. [Figure 2] FIG. 1 is a cross-sectional view of a CMP apparatus. [Figure 3] FIG. 2 is a cross-sectional view showing an example of the structure of a wafer. [Figure 4] FIG. 2 is a schematic diagram showing an example of the relationship between the thickness of the underlayer and the measured value of the thickness of the polishing layer. [Figure 5] FIG. 10 is a schematic diagram showing an example of a method for setting waveform model generation parameters. [Figure 6] FIG. 10 is a schematic diagram showing an example of a database of model waveforms of reflection spectra of polishing layer thicknesses corresponding to the thickness of the base layer according to the embodiment. [Figure 7] FIG. 10 is a schematic diagram showing an example of a method for acquiring a reflected light spectrum. [Figure 8] 1 is a flowchart showing an example of a polishing method using the CMP apparatus according to the embodiment. [Figure 9] FIG. 10 is a schematic diagram showing an example of a database of model waveforms of reflection spectra for polishing layer thicknesses of comparative examples. DETAILED DESCRIPTION OF THE INVENTION

[0016] In the semiconductor manufacturing process, the CMP process, which involves forming wiring patterns and shallow trench isolation on film-processed wafers, requires precise control of the final film thickness by stopping polishing at the exact polishing endpoint. The control of film thickness requires detecting the polishing endpoint in real time and ensuring a consistent film thickness.

[0017] FIG. 1 is a perspective view of the main part of a CMP apparatus 1 according to an embodiment, and FIG. 2 is a cross-sectional view of the CMP apparatus 1. As shown in FIG. 1, the CMP apparatus 1 includes a platen 2, a motor 3, a drive mechanism 4, a polishing pad 5, a polishing head 6, and a processing device 100. Note that the CMP apparatus 1 may include components other than the platen 2, the motor 3, the drive mechanism 4, the polishing pad 5, the polishing head 6, and the processing device 100, or may not include at least one of these components.

[0018] The platen 2 is formed, for example, in a disk shape. The platen 2 is attached on a drive mechanism 4 using a motor 3. A polishing pad 5 is attached to the upper surface of the platen 2. The polishing head 6 rotates around the rotation axis of the polishing head 6. The polishing head 6 holds a wafer W, which is to be polished, on its lower surface. The polishing head 6 is, for example, in a disk shape with a smaller diameter than the platen 2. The polishing head 6 is, for example, disposed above the platen 2 at a position displaced from the rotation axis of the platen 2.

[0019] The processing device 100 is connected to each part of the CMP device 1. The processing device 100 may also be connected to devices other than the CMP device 1. The processing device 100 controls each part of the CMP device 1 and / or each part of devices other than the CMP device 1, transmits and receives information between each part of the CMP device 1 and / or each part of devices other than the CMP device 1, records information, and performs various calculations based on acquired information. The processing device 100 may also be provided as a separate entity from the CMP device 1.

[0020] The processing device 100 includes a control unit 110, a detection unit 120, an arithmetic processing unit 130, and a storage unit 140. The processing device 100 may include components other than the control unit 110, the detection unit 120, the arithmetic processing unit 130, and the storage unit 140, or may not include at least one of these components.

[0021] The control unit 110 controls, for example, each part of the CMP apparatus 1 and each part of other apparatuses other than the CMP apparatus 1.

[0022] The detection unit 120 performs various detections. The detection unit 120 performs various detections based on, for example, information and data output from each unit of the CMP apparatus 1 and each unit of other devices other than the CMP apparatus 1. The arithmetic processing unit 130 performs various calculations and processes. The arithmetic processing unit 130 performs various calculations and processes based on, for example, information exchanged between each unit of the CMP apparatus 1 and each unit of other devices other than the CMP apparatus 1. The memory unit 140 stores information exchanged between each unit of the CMP apparatus 1 and each unit of other devices other than the CMP apparatus 1, detection results by the detection unit 120, and calculation results calculated by the arithmetic processing unit 130. The detection unit 120, the arithmetic processing unit 130, and the memory unit 140 may each be provided as a device separate from the CMP apparatus 1 and the processing device 100. For example, the detection unit 120, the arithmetic processing unit 130, and the memory unit 140 may each be a PC (personal computer) or a server separate from the CMP apparatus 1 and the processing device 100.

[0023] The polishing process is carried out by dripping abrasive slurry from a slurry supply device (not shown) onto the upper surface of the polishing pad 5, lowering the polishing head 6 and wafer W to a height where they come into contact with the upper surface of the polishing pad 5, and then rotating the platen 2 and polishing head 6 to polish the surface (lower surface) of the wafer W to be polished.

[0024] 2, platen 2 has a hole 7 penetrating vertically, polishing pad 5 has a hole 8 penetrating vertically, and a transparent window 9 is embedded in hole 8 of polishing pad 5. With each rotation of platen 2, window 9 passes over the underside of wafer W attached to polishing head 6. On the underside of platen 2 and vertically below polishing head 6, an optical sensor head 10 of an optical film thickness measuring device is disposed.

[0025] The optical film thickness measurement device includes a light source (not shown), a spectroscope (not shown), and an optical sensor head 10 connected to the light source and spectroscope. The optical film thickness measurement device may be included in the CMP device 1 or may be provided separately from the CMP device 1. The optical sensor head 10, the light source, and the spectroscope are attached to the platen 2 and rotate integrally with the platen 2 and the polishing pad 5. The spectroscope is connected to, for example, the processing device 100. The spectroscope may be connected to at least one of the control unit 110, the detection unit 120, and the arithmetic processing unit 130 of the processing device 100. The processing device 100 may be included in the optical film thickness measurement device or may be provided separately from the optical film thickness measurement device. The control unit 110, the detection unit 120, the arithmetic processing unit 130, and the memory unit 140 may each be included in the optical film thickness measurement device or may be provided separately from the optical film thickness measurement device.

[0026] Light emitted from the light source is transmitted to the optical sensor head 10 and directed from the optical sensor head 10 to the surface of the wafer W. The light is reflected from the surface of the wafer W, and the reflected light from the surface of the wafer W is received by the optical sensor head 10 and sent to a spectrometer.

[0027] The spectrometer separates the reflected light according to wavelength and measures the intensity of the reflected light at each wavelength. The spectrometer, for example, transmits the measurement data of the reflected light intensity to a processing device 100, for example, a detection unit 120. The intensity of the reflected light can also be expressed as a relative value such as reflectance or relative reflectance. The spectrum obtained by the spectrometer uses a physical quantity proportional to the wavelength of the electromagnetic wave or the energy of light (e.g., wave number, frequency, electron volt) on the horizontal axis, and light intensity or a physical quantity derived from the intensity (degree of polarization) on the vertical axis.

[0028] The detection unit 120 generates a reflected light spectrum based on the reflected light intensity measurement data, and determines, detects, or measures the thickness of the polishing layer based on the generated reflected light spectrum. The reflected light spectrum is expressed as a line graph (i.e., a spectral waveform) showing the relationship between the wavelength and intensity of the reflected light. The detection unit 120 also generates the reflected light spectrum and determines the thickness (or film thickness) of the polishing layer formed on the wafer surface based on the reflected light spectrum. The detection unit 120 determines the polishing endpoint of the polishing layer based on the determined, detected, or measured thickness of the polishing layer, and stops polishing or changes the polishing conditions. In other words, the detection unit 120 determines the polishing endpoint of the polishing layer based on the determined, detected, or measured reflected light spectrum, and stops polishing or changes the polishing conditions.

[0029] The detection unit 120 acquires a reflected light spectrum in situ during polishing, and refers to a database 30 in which model waveforms 31 of reflected light spectra for thicknesses (or film thicknesses) that serve as a reference for determining the end point of the polishing layer, as shown in Fig. 6 described below, are stored, and determines, detects, or measures the thickness of the polishing layer when the acquired reflected spectrum matches or resembles a predetermined model waveform 31 in the database 30, or when the acquired reflected spectrum matches or approximates a model waveform 31 for a thickness equal to or less than a predetermined thickness. The detection unit 120 is equipped with, for example, a program that detects the end point of polishing based on the determination, detection, or measurement results and stops polishing or changes the polishing conditions.

[0030] For example, the detection unit 120 acquires a reflected light spectrum in-situ during polishing, and when it determines by referring to the database 30 that the thickness of the polished layer matches a predetermined target thickness (hereinafter referred to as the target thickness or the target film thickness), when it determines that the thickness is close to the target thickness, or when it determines that the thickness is equal to or less than the target thickness, it stops polishing of the wafer W by the CMP apparatus 1 via the control unit 110. In other words, the detection unit 120 stops polishing of the wafer W by the CMP apparatus 1, for example, when the reflected light spectrum acquired during polishing matches or resembles a model waveform of the target thickness, or when the reflected light spectrum acquired during polishing matches or resembles a model waveform of a thickness equal to or less than the target thickness.

[0031] The arithmetic processing unit 130, for example, sets predetermined parameters and generates a database 30 of model waveforms of reflected light spectra of the thickness (or film thickness) of the polishing layer corresponding to the thickness (or film thickness) of the base layer located below the polishing layer. Note that a portion other than the arithmetic processing unit 130 may generate this database 30. For example, the control unit 110, the detection unit 120, the optical film thickness measurement device, and / or the CMP apparatus 1 may generate this database 30, or a device or portion other than the control unit 110, the detection unit 120, the optical film thickness measurement device, and the CMP apparatus 1 may generate this database 30.

[0032] The arithmetic processing unit 130 sets parameters (hereinafter sometimes referred to as waveform model generation parameters) such as information on the material of the polishing layer, information on the material of the base layer located below the polishing layer, the material of the substrate, the range of the expected thickness (or expected film thickness) of the polishing layer (hereinafter sometimes referred to as the polishing layer thickness range or the polishing layer film thickness range), the range of the expected thickness (or expected film thickness) of the base layer (hereinafter sometimes referred to as the base layer thickness range or the base layer film thickness range), and the interval of the layer thickness to be calculated (hereinafter sometimes simply referred to as the thickness interval or the film thickness interval). The material information includes various information such as the type of material, the name of the material, the characteristics of the material, the refractive index of the material, and other physical properties. The arithmetic processing unit 130 sets, for example, the polishing layer thickness range and the base layer thickness range as variable parameters among the waveform model generation parameters. The polishing layer thickness range is, for example, a group of polishing target film data including at least an initial film thickness value (DS1) at the start of the polishing operation and a target film thickness value (DS2) after the polishing operation as film thickness values, and further including film thickness values ​​between them. Furthermore, the underlayer thickness range is a group of underlayer data including a reference underlayer thickness value (DI1) and an error thickness value (DIx) within an error range centered on this reference thickness value. The polishing layer thickness range, underlayer thickness range, and thickness interval are preferably adjusted according to the processing capacity of the processing apparatus 100 or the device or equipment that generates the database 30. The thickness interval may also be set in advance as an appropriate reference interval for detecting the polishing endpoint depending on the target material.

[0033] The arithmetic processing unit 130 sets waveform model generation parameters based on values ​​of the waveform model generation parameters input, for example, on an input screen 20 such as that shown in FIG. 5 (described later). The arithmetic processing unit 130 generates a database 30 of model waveforms of the reflection spectrum of a polishing layer of a predetermined material at each thickness corresponding to the base layer of each predetermined material thickness by simulating each model waveform of the reflection spectrum of a polishing layer of a predetermined material at each thickness according to the set waveform model generation parameters based on data on the material (information) and thickness of the base layer and data on the material of the substrate. The arithmetic processing unit 130 may be capable of adding and setting materials of layers other than the polishing layer, base layer, and substrate, and ranges of expected thicknesses of those layers. The arithmetic processing unit 130 may also add information (or refractive index data, etc.) on the predetermined material.

[0034] The arithmetic processing unit 130 may store the generated database 30 in the storage unit 140. Note that the arithmetic processing unit 130 may store the generated database 30 in a storage unit other than the storage unit 140.

[0035] The memory unit 140 stores a program for generating a spectrum of reflected light and determining the thickness (or film thickness) of the polishing layer (or measurement layer) formed on the wafer surface from the spectrum of reflected light, information (or data) used for determining the film thickness, such as database 30, measured values ​​of the detected polishing layer thickness, and various other information (or data).

[0036] FIG. 3 is a cross-sectional view showing an example of the structure of the wafer W. The wafer W includes, for example, a polishing layer (or measurement layer) W1, which is the uppermost layer to be polished and whose thickness is to be measured, a base layer W2 provided below the polishing layer W1, and a substrate W3 provided below the base layer W2. In the example shown in FIG. 3, the base layer W2 is in contact with the underside of the polishing layer W1, and the substrate W3 is in contact with the underside of the base layer W2. Note that, although the base layer W2 is a single layer in the example shown in FIG. 3, it may be multiple layers. For example, the wafer W is an SOI substrate. In this case, the polishing layer W1 is formed of a material containing lithium tantalate (LT), which can be used for a broadband, the base layer W2 includes at least one layer formed of a material containing SiO2, and the substrate W3 is formed of a material containing Si.

[0037] 4 is a schematic diagram showing an example of the relationship between the thickness of the base layer W2 and the measured value of the thickness of the polishing layer W1. In FIG. 4, for example, the actual thickness WT1 of the polishing layer W1 is constant relative to the thickness of the base layer W2. In the example shown in FIG. 4, the measured thickness WTm of the polishing layer W1, measured without considering the influence of the base layer W2 as in the conventional method, is detected from the reflected light spectrum used to measure the thickness of the polishing layer W1. In the example shown in FIG. 4, the measured thickness WTm of the polishing layer W1 increases as the thickness of the base layer W2 increases. In other words, when the thickness of the polishing layer W1 is measured without considering the influence of the base layer W2, an error may occur in the measured thickness of the polishing layer W1. This is thought to be because the reflected light spectrum used to measure the thickness of the polishing layer W1 is affected by the base layer W2. Furthermore, the thickness of the base layer W2 is not necessarily the same as the designed thickness. For example, if the designed thickness of the base layer W2 is 500 nm, the actual thickness of the base layer W2 may be manufactured to a thickness ranging from 500 nm to 520 nm.

[0038] As mentioned above, the inventors of the present application have discovered that if the thickness of the polishing layer W1 is determined based on a database of model waveforms of reflected light spectra for the thickness of the polishing layer W1 with the thickness of the base layer W2 fixed, as shown in database 300 of model waveforms of reflected light spectra generated by simulation of the comparative example in Figure 9, and the measured reflected light spectrum, the resulting value may differ from the actual thickness of the polishing layer W1.

[0039] FIG. 5 is a schematic diagram showing an example of a method for setting waveform model generation parameters. The processing device 100 has, for example, an input screen 20 for inputting values ​​of waveform model generation parameters. Note that the input screen 20 may be provided in a device or part other than the processing device 100. Furthermore, the processing device 100 does not need to have the input screen 20, and may have another screen, device, equipment, function, etc. as long as it is possible to set or input values ​​of waveform model generation parameters.

[0040] The processing device 100 (e.g., the calculation processing unit 130) has a means for displaying material items 21 and specification items 22, which are information necessary for the simulation, and a program for realizing the input function in order to realize the input function of information about the wafer W.

[0041] The material item 21 includes, for example, the type of material of the polishing layer W1, the type of material of the base layer W2, and the type of material of the substrate W3. In the example shown in Fig. 5, in the material item 21, the input value 23 for the type of material of the polishing layer W1 is input as LT, the input value 24 for the type of material of the base layer W2 is input as SiO2, and the input value 25 for the type of material of the substrate W3 is input as Si. Note that the material item 21 may also have an input field for inputting the refractive index of the material.

[0042] Specification items 22 are the polishing layer thickness range of polishing layer W1, the base layer thickness range of base layer W2, and the thickness interval between polishing layer W1 and base layer W2. The input value 26 for the thickness range of polishing layer W1 (polishing layer thickness range) is 100 to 900 nm, the input value 27 for the thickness range of base layer W2 (base layer thickness range) is 400 to 600 nm, and the input value 28 for the thickness interval between polishing layer W1 and base layer W2 is 5 nm. The polishing layer thickness range, base layer thickness range, and thickness interval may be changed depending on the polishing accuracy and the processing capacity of the processing device 100.

[0043] The processing device 100 has, for example, a program for executing a function of generating a model waveform 31 of the reflected light spectrum by simulation for the material and / or refractive index input in the material item 21 and the thickness range (polishing layer thickness range and base layer thickness range) input in the specification item 22, and a program for executing a function of storing the simulated model waveform 31 as a database 30.

[0044] FIG. 6 is a schematic diagram showing an example of a database 30 of model waveforms 31 of reflection spectra of polishing layer thicknesses corresponding to the thicknesses of the foundation layers according to this embodiment. In Figure 6, the thickness of the polishing layer varies along the horizontal axis, and the thickness of the base layer varies along the vertical axis. In Figure 6, the thickness interval between each model waveform 31 is 5 nm. That is, the difference in the thickness of the polishing layer between two adjacent model waveforms 31 on the horizontal axis is 5 nm. Also, the difference in the thickness of the base layer between two adjacent model waveforms 31 on the vertical axis is 5 nm.

[0045] When the thickness range (polishing layer thickness range) of the polishing layer W1 is input as 100 to 900 nm, the thickness range (base layer thickness range) of the base layer W2 is input as 400 to 600 nm, and the thickness interval is 5 nm, the processing device 100 generates a database 30 of model waveforms 31 of the reflected light spectrum of the polishing layer W1 with 160 rows and 40 columns by simulation based on these waveform model generation parameters.

[0046] Furthermore, when manufacturing the underlayer W2 with a design value of 500 nm, the underlayer W2 may be manufactured in the range of 500 to 520 nm, so the thickness interval of the underlayer W2 needs to be 10 nm or less, which is at least half the error range of the thickness of the underlayer W2, 20 nm, and it is preferable to set it to, for example, 5 nm.

[0047] The processing device 100 refers to the database 30 and selects one model waveform 31 from the database 30 that matches or is most similar to the reflected light spectrum acquired in situ during polishing, i.e., the model waveform 31 with the highest similarity, or selects a model waveform 31 that corresponds to a thickness equal to or less than the target thickness. For example, the processing device 100 selects the model waveform 31 enclosed in a box in FIG. 6. The thickness of the polishing layer W1 at this time is determined from the row position to which the selected model waveform 31 belongs. The thickness of the base layer W2 can be determined from the column position to which the selected model waveform 31 belongs.

[0048] The degree of agreement or similarity between the model waveform 31 and the acquired reflected light spectrum is determined by numerical calculation processing such as a method of calculating a correlation coefficient that indicates the strength of the relationship between the two waveforms, or a dynamic time warping (DTW) method that calculates the distance between each point of the two waveforms in a brute force manner to find the combination that results in the smallest distance.

[0049] In other words, the processing device 100 has a program for realizing the function of numerically calculating the similarity between the model waveform 31 stored in the database 30 shown in Figure 6 and the reflected light spectrum acquired during polishing (in-situ), and the function of selecting one model waveform 31 with the highest processed similarity, and a program for realizing the function of determining the film thickness of the polishing layer W1 from the selected model waveform 31.

[0050] FIG. 7 is a schematic diagram showing an example of a method for acquiring a reflected light spectrum. The film thickness at a measurement point on the wafer W is determined based on the spectrum of light reflected from that measurement point. Each time the platen 2 rotates once, the optical sensor head 10 directs light to a predetermined measurement point on the wafer W and acquires or measures the spectrum of light reflected from that measurement point.

[0051] The spectrometer separates the reflected light according to wavelength and obtains the reflected light intensity measurement data for each wavelength as a reflected light spectrum. Although the transparent window 9 (see Figure 2) affects the intensity of the reflected light, this effect can be avoided by measuring the standard reflection state in advance and calibrating.

[0052] The acquired reflected light spectrum is represented as a line graph (i.e., a spectral waveform) showing the relationship between the wavelength and intensity of light, similar to the model waveform 31, with the horizontal axis representing the wavelength of the reflected light and the vertical axis representing the reflectance (%) derived from the intensity of the reflected light. The irradiation of the wafer W with light, reception of the reflected light, measurement of the intensity of the reflected light, and generation of the reflected light spectrum based on the intensity measurement of the reflected light are repeated each time the platen 2 rotates in situ during polishing.

[0053] FIG. 8 is a flowchart showing an example of a polishing method using the CMP apparatus 1 according to this embodiment. The processing device 100 receives input information (values ​​of waveform model generation parameters) to generate a database 30 of model waveforms 31 of the reflected light spectrum of a polishing layer W1 of a specified material with each film thickness corresponding to a base layer of a specified material with each thickness through simulation (step S1).

[0054] The processing device 100 generates a database 30 of model waveforms 31 of the reflection spectrum of the polishing layer W1 of each film thickness of a predetermined material corresponding to each thickness of the base layer of the predetermined material through simulation based on the information (values ​​of the waveform model generation parameters) input in step S1 on the input screen 20. For example, the processing device 100 calculates the model waveforms 31 of each film thickness of the polishing layer W1 at film thickness intervals within the film thickness range of the polishing layer W1 and base layer W2 in the specification item 22 based on the refractive indexes of the material types of the polishing layer W1, base layer W2, and substrate W3 input in the material item 21, and generates the database 30 of the model waveforms 31 of the reflection spectrum of the polishing layer W1.

[0055] The processing device 100 stores the generated database (step S3). For example, the processing device 100 stores the generated database in the storage unit 140.

[0056] Next, the processing device 100 measures the reflected light intensity measurement data in-situ during polishing to obtain the reflected light spectrum (step S4). For example, the processing device 100 measures the reflected light intensity measurement data in-situ during polishing to obtain the reflected light spectrum of the polishing layer W1.

[0057] The processing device 100 determines the thickness of the polishing layer W1 by referring to the database 30 shown in Fig. 6 (step S5).

[0058] Next, the processing device 100 determines the polishing end point of the polishing layer W1 based on the determined thickness of the polishing layer W1, and stops polishing or changes the polishing conditions (step S6). Preferably, the processing device 100 stops polishing when it determines that the determined thickness of the polishing layer W1 matches, is close to, or is equal to or less than the target thickness.

[0059] According to this embodiment, the CMP apparatus 1 (e.g., processing apparatus 100) refers to a database 30 in which the polishing layer thickness range of the polishing layer W1 and the base layer thickness range of the base layer W2 are variable parameters, and determines the film thickness of the polishing layer W1 based on model waveforms 31 of the reflected light spectra of the polishing layer W1 stored in the database 30 and the reflected light spectra acquired in situ during polishing. Therefore, according to the CMP apparatus 1 (processing apparatus 100) of this embodiment, even if the base layer W2 is not manufactured to have the thickness as designed, the thickness of the polishing layer W1 can be detected with high accuracy.

[0060] Furthermore, since the referenced database 30 can be generated by computational simulation if the materials of the polishing layer W1 and the underlayer W2 are known, there is no need to measure the initial film thickness and the film thickness after polishing in advance, and therefore it is easy to increase the number of model waveforms 31 in the referenced database 30 as shown in Fig. 6. As a result, the CMP apparatus 1 (processing apparatus 100) according to this embodiment can be applied to wafers W having more than two layers, namely the polishing layer W1 and the underlayer W2, as an example of a configuration, and it is also possible to easily generate a database 30 of the film thickness of the first layer when the film thickness of the second layer and subsequent layers is changed.

[0061] That is, the CMP apparatus 1 according to this embodiment can be applied to cases where the number of layers is increased, as long as the processing capacity of the PC constituting the processing apparatus 100 is sufficient to process the multi-column database 30. The simulation can be performed whenever information about the wafer W (values ​​of the waveform model generation parameters) is input and preparations for polishing the polishing layer W1 are made in the CMP apparatus 1, so it can be performed for each individual process, and the thickness of the polishing layer W1 can be controlled with higher precision under the conditions for each process.

[0062] As described above, the CMP apparatus 1 (processing apparatus 100) according to this embodiment can easily improve the accuracy of measuring the thickness of the polishing layer among the multiple layers stacked on the wafer surface.

[0063] In step S4 of the flowchart shown in FIG. 8, the reflected light intensity measurement data may occasionally contain abnormal values ​​measured as singular points, but this can be addressed by calculating the moving average of the intensity measurement data to smooth them out.

[0064] The above embodiment is merely an example, and the present invention may be modified and implemented as appropriate without departing from the spirit and scope of the present invention. Furthermore, a process described as being performed by one device may be shared and executed by multiple devices. Alternatively, a process described as being performed by different devices may be executed by a single device.

[0065] The processing device 100 is a computer system, and the hardware configuration (server configuration) by which each function is realized can be flexibly changed. The present invention can also be realized by supplying a computer program that implements the functions described in the above embodiments to a computer, and having one or more processors in the computer read and execute the program.

[0066] That is, the processing device 100 may be a plurality of servers connected by a communication network such as the Internet or a local area network. In one embodiment, the database 30 shown in Figure 6 may be provided in a data server located remotely from the CMP device 1 (see Figure 1). [Explanation of symbols]

[0067] 1 CMP equipment 2 Platen 3 motors 4. Drive mechanism 5 polishing pads 6 polishing heads 7 holes 8 holes 9 Windows 10 Optical sensor head 20 Input screen 30 databases 100 Processing equipment 110 control section 120 Detector 130 Processing unit 140 Storage section 300 databases

Claims

1. 1. A CMP apparatus for polishing a polishing layer of a wafer having a polishing layer and at least one foundation layer located below the polishing layer, comprising: a database including model waveforms of reflected light spectra of the polishing layer at each thickness corresponding to each thickness of the base layer; a detection unit that determines the thickness of the polishing layer based on the database and a first reflected light spectrum detected during polishing of the wafer; a processing unit that generates the database based on information on the material of the polishing layer, information on the material of the base layer, a first thickness range of the polishing layer, a second thickness range of the base layer, and thickness intervals between the polishing layer and the base layer.

2. 2. The CMP apparatus according to claim 1, wherein the detection unit stops polishing of the wafer when the first reflected light spectrum matches or resembles a model waveform in the database.

3. 2. The CMP apparatus according to claim 1, wherein the detection unit stops polishing of the wafer when the first reflected light spectrum matches or resembles a model waveform corresponding to a thickness equal to or less than a target thickness in the database.

4. the first thickness range includes an initial thickness value of the polishing layer at the start of a polishing operation and a target thickness value of the polishing layer after the polishing operation is completed; 4. The CMP apparatus according to claim 2, wherein the second thickness range includes an error range that occurs when manufacturing the underlayer.

5. A polishing method applied to a CMP apparatus for polishing a polishing layer of a wafer having a polishing layer and at least one foundation layer located below the polishing layer, comprising: generating a database based on information about the material of the polishing layer, information about the material of the base layer, a first thickness range of the polishing layer, a second thickness range of the base layer, and thickness intervals between the polishing layer and the base layer; The polishing method further comprises determining a thickness of the polishing layer based on the database and a first reflected light spectrum detected during polishing of the wafer.

Images