Goodness of fit in spectrographic monitoring of a substrate during processing

0001This specification generally relates to spectroscopic monitoring of a substrate during chemical mechanical polishing.

0002Integrated circuits are typically formed on a substrate by sequential deposition of conductive, semi-conductive, or insulating layers on a silicon wafer. One manufacturing step involves depositing a filler layer over an uneven surface and flattening the filler layer. For certain applications, the filler layer is flattened until the top surface of the patterned layer is exposed. For example, a conductive filler layer can be deposited on the patterned insulating layer to fill a trench or hole in the insulating layer. After flattening, some of the conductive layers left between the elevated patterns of insulating layers form vias, plugs, and wiring that provide conductive paths between the thin film circuits on the substrate. For other applications such as oxide polishing, the filler layer is flattened until a predetermined thickness is left above the non-flat surface. In addition, substrate surface flattening is typically required for photolithography.

0003Chemical mechanical polishing (CMP) is one recognized method of flattening. This flattening method typically requires mounting the substrate on a carrier or polishing head. The exposed surface of the substrate is typically placed towards a rotating abrasive disc pad or belt pad. The polishing pad can be either a standard pad or a fixed abrasive pad. Standard pads have a durable, roughened surface, whereas fixed abrasive pads have abrasive particles retained in the containment medium. The carrier head provides a controllable load on the substrate to push the substrate towards the polishing pad. A polishing solution, such as a slurry, comprising abrasive particles is typically supplied to the polishing pad on the surface.

0004One question in CMP is determining whether the polishing process is over, that is, whether the substrate layer has been flattened to the desired flatness or thickness, or when the desired amount of material has been removed. It is to be. Overpolishing (excessive removal) of the conductive layer or film leads to an increase in circuit resistance. On the other hand, underpolishing (too little removal) of the conductive layer leads to an electrical short circuit. Fluctuations in slurry composition, polishing pad condition, relative speed between polishing pad and substrate, and load on the substrate, for example, removal rate of substances between substrates or between different zones of a single substrate. May cause fluctuations in. Moreover, variations in the initial thickness of the substrate layer, between substrates, or between different zones of a single substrate can cause the amount of variation that needs to be removed before reaching the target thickness. These fluctuations cause fluctuations in the time required to reach the polishing end point. Therefore, the polishing end point cannot be reliably determined simply as a function of polishing time.

0005In one general aspect, the computer mounting method is to obtain a series of current spectra using an in-situ optical monitoring system, where each current spectrum from the series of current spectra is polished. It is a spectrum of light reflected from a substrate with an outermost layer and at least one underlying layer, which is available and compares each current spectrum to multiple reference spectra from the first reference spectrum library. To generate the first series of first best match reference spectra, determine the first best match reference spectrum, determine the first fit for the first series, and first. Includes determining the polishing end point based on a series of 1 and the first degree of conformity.

0006In another general aspect, the computer mounting method is to obtain a series of current spectra using an in-situ optical monitoring system, where each current spectrum from the series of current spectra polishes. Obtain and compare each current spectrum to multiple reference spectra from multiple reference spectrum libraries, which are spectra of light reflected from a substrate having an outermost layer receiving and at least one underlying layer. And to determine which library gives the best fit to the current spectral sequence, and to determine the polishing endpoint based on the library that gives the best fit to the current spectral sequence and the current spectral sequence. including.

0007Implementation examples of either of these two methods can include one or more of the features below. A second best-match reference to generate a second series of second best-match reference spectra, which allows the current spectrum to be compared against multiple reference spectra from the second best-match spectrum library. The spectrum can be determined, the second goodness of fit can be determined for the second series, and the end point can be the first series, the second series, the first goodness of fit, or the second. Can be determined based on the goodness of fit of. Determining the polishing end point determines whether the first best match reference spectrum points to the end point, and if so, is the first goodness of fit better than the second goodness of fit? It can include deciding whether or not, and if so, declaring an end point. Determining the polishing end point determines whether the second best match reference spectrum points to the end point, and if so, is the second goodness of fit better than the first goodness of fit? It can include deciding whether or not, and if so, declaring an end point. To generate a series of first index values, the first index value can be determined for each first best match reference spectrum, and determining the second goodness of fit can be the second index. To generate a series of values, it can include determining a second index value for each second best match reference spectrum. The first function can be fitted to a series of first index values and the second function can be fitted to a series of second index values. The first function and the second function can be linear functions. Determining the goodness of fit can include determining the goodness of fit of a series of first index values to a first function, and determining the second goodness of fit is a second. It can include determining the goodness of fit of a series of second index values to a function. Determining the first goodness of fit is the first index Determining the sum of squares of the differences between the sequence of values and the first function, and determining the second goodness of fit can include determining the sequence of second index values and the second function. It can include determining the sum of squares of the differences between and. A series of first index values can form a first index mark, and a series of second index values can form a second index mark. Determining whether the first best match reference spectrum points to an end point can include determining whether the index of the first best match reference spectrum is the target index. Determining whether the second best match reference spectrum points to an end point can include determining whether the index of the second best match reference spectrum is the target index. It can be determined whether the first goodness of fit for the first series is better than the second goodness of fit for the second series. Multiple reference spectra from the first reference spectrum library can represent a substrate with a first thickness underlayer, and multiple reference spectra from the second reference spectrum library have different second thicknesses. A substrate having a base layer can be represented. Determining the first best-matched reference spectrum can include determining which reference spectrum from the first reference spectrum library has the smallest difference from the current spectrum and is the second best. Determining the match reference spectrum can include determining which reference spectrum from the second reference spectrum library has the smallest difference from the current spectrum. Dex marks can be formed. Determining whether the first best match reference spectrum points to an end point can include determining whether the index of the first best match reference spectrum is the target index. Determining whether the second best match reference spectrum points to an end point can include determining whether the index of the second best match reference spectrum is the target index. It can be determined whether the first goodness of fit for the first series is better than the second goodness of fit for the second series. Multiple reference spectra from the first reference spectrum library can represent a substrate with a first thickness underlayer, and multiple reference spectra from the second reference spectrum library have different second thicknesses. A substrate having a base layer can be represented. Determining the first best-matched reference spectrum can include determining which reference spectrum from the first reference spectrum library has the smallest difference from the current spectrum, and the second best. Determining the match reference spectrum can include determining which reference spectrum from the second reference spectrum library has the smallest difference from the current spectrum. Dex marks can be formed. Determining whether the first best match reference spectrum points to an end point can include determining whether the index of the first best match reference spectrum is the target index. Determining whether the second best match reference spectrum points to an end point can include determining whether the index of the second best match reference spectrum is the target index. It can be determined whether the first goodness of fit for the first series is better than the second goodness of fit for the second series. Multiple reference spectra from the first reference spectrum library can represent a substrate with a first thickness underlayer, and multiple reference spectra from the second reference spectrum library have different second thicknesses. A substrate having a base layer can be represented. Determining the first best-matched reference spectrum can include determining which reference spectrum from the first reference spectrum library has the smallest difference from the current spectrum and is the second best. Determining the match reference spectrum can include determining which reference spectrum from the second reference spectrum library has the smallest difference from the current spectrum. Multiple reference spectra from the second reference spectrum library can represent substrates with different second thickness underlying layers. Determining the first best-matched reference spectrum can include determining which reference spectrum from the first reference spectrum library has the smallest difference from the current spectrum, and the second best. Determining the match reference spectrum can include determining which reference spectrum from the second reference spectrum library has the smallest difference from the current spectrum. Multiple reference spectra from the second reference spectrum library can represent substrates with different second thickness underlying layers. Determining the first best-matched reference spectrum can include determining which reference spectrum from the first reference spectrum library has the smallest difference from the current spectrum, and the second best. Determining the match reference spectrum can include determining which reference spectrum from the second reference spectrum library has the smallest difference from the current spectrum.

0008In general, another aspect of the subject matter described herein can be embodied in computer implementation methods involving receiving a first series of current spectra of reflected light from a first zone of a substrate. .. A second sequence of the current spectrum of reflected light from the second zone of the substrate can be received. To generate a first series of best-matched reference spectra, it is possible to compare each current spectrum from the first series of current spectra against multiple reference spectra from the first reference spectrum library. it can. To generate a second series of best-matched reference spectra, it is possible to compare each current spectrum from the second series of current spectra against multiple reference spectra from the second reference spectrum library. it can. The second reference spectrum library may differ from the first reference spectrum library. Other embodiments of this aspect include corresponding systems, devices, and computer program products.

0009These and other embodiments may optionally include one or more of the following features: The first reference spectrum library and the second reference spectrum library can be predetermined.

0010In another aspect, a computer program product tangibly embodied in a computer-readable medium can operate to cause a data processor to perform an operation that includes the steps of the method described above.

0011In general, one aspect of the subject matter described herein can be embodied in computer implementation methods involving receiving a first sequence of current spectra of reflected light from a first zone of a substrate. A second sequence of the current spectrum of reflected light from the second zone of the substrate can be received. Each present from the first series of current spectra, for the first plurality of reference spectra from the first plurality of reference spectrum libraries, in order to generate multiple first series of best-matched reference spectra. Spectrums can be compared. Each present from the second series of current spectra, for the second plurality of reference spectra from the second plurality of reference spectrum libraries, in order to generate multiple second series of best-matched reference spectra. Spectrums can be compared. It is possible to determine a plurality of goodness of fit for a plurality of first series of best match reference spectra. It is possible to determine multiple goodness of fit for multiple second series of best match reference spectra. Other embodiments of this aspect include corresponding systems, devices, and computer program products.

0012These and other embodiments may optionally include one or more of the following features: Some of the first plurality of reference spectrum libraries and the second plurality of reference spectrum libraries may be the same. The first plurality of reference spectrum libraries and the second plurality of reference spectrum libraries may all be the same. Neither the first plurality of reference spectrum libraries nor the second plurality of reference spectrum libraries may be the same.

0013Generating multiple first series of best-matched reference spectra can compare each current spectrum from the first series of current spectra to multiple reference spectra from the first reference spectrum library. A second, determining the first intermediate fit and comparing each current spectrum from the first series of current spectra against multiple reference spectra from the second reference spectrum library. The first is based on determining the intermediate fit, comparing the first intermediate fit to the second intermediate fit, and comparing the first intermediate fit to the second intermediate fit. Includes first selection of one of the reference spectrum libraries or the second reference spectrum library of, and determination of the first series of best-matched reference spectra based on the first selection.

0014Generating multiple second series of best-matched reference spectra can compare each current spectrum from the second series of current spectra to multiple reference spectra from the first reference spectrum library. A fourth, determining the third intermediate fit and comparing each current spectrum from the second series of current spectra against multiple reference spectra from the second reference spectrum library. The first is based on determining the intermediate fit, comparing the third intermediate fit to the fourth intermediate fit, and comparing the third intermediate fit to the fourth intermediate fit. Includes the second selection of one of the reference spectrum libraries or the second reference spectrum library of, and the determination of the second series of best-matched reference spectra based on the second selection.

0015The first and second choices can be determined within a predetermined period of polishing. The predetermined period of polishing can include the first 20 seconds of polishing. The method determines the first polishing end point for the first zone based on the first sequence of the best match reference spectrum and the corresponding first goodness of fit, and the second sequence of the best match reference spectrum. And determining the second polishing end point for the second zone based on the corresponding second goodness of fit can further be included.

0016As used herein, the term substrate may include, for example, a product substrate (which may include, for example, multiple memory or processor dies), a test substrate, a bare substrate, or a gating substrate. it can. The substrate may be at various stages of integrated circuit manufacturing, for example, the substrate may be a bare wafer, or may include one or more deposited layers and / or patterned layers. The term substrate can include circular discs and rectangular sheets.

0017Potential advantages of the implementation examples of the present invention can include: The endpoint detection system may not be sensitive to variations in the underlying layer or pattern between substrates or between zones on a single substrate, and therefore the desired for one substrate or for each zone of one substrate. The reliability of the end point detection system that detects the end point can be improved. As a result, the thickness uniformity between the wafers and inside the wafer can be improved.

0018Details of one or more embodiments of the present invention will be described in the accompanying drawings and in the description below. Other features, aspects, and advantages of the present invention will become apparent from the description, drawings, and claims.

0019<figref num="1">It is sectional drawing of a part of a substrate.</figref><figref num="2">It is sectional drawing which illustrates an example of a polishing apparatus.</figref><figref num="3">It is a bird's-eye view of an example rotating platen exemplifying the place of in-situ measurement.</figref><figref num="4">It is a schematic diagram illustrating an example index trace from a spectroscopic monitoring system showing good data fitting.</figref><figref num="5">It is a schematic diagram illustrating an example index trace from a spectroscopic monitoring system showing poor fitting.</figref><figref num="6">It is a flow chart of one mounting example which determines a polishing end point.</figref><figref num="7">It is a graph of an example of polishing progress vs. time relating to the process of adjusting the polishing rate.</figref>

0020Similar reference numbers and symbols in various drawings refer to similar elements.

0021During polishing, substrates with different patterns and different underlying layer thicknesses may pass through the polishing device. When attempting to use spectroscopic monitoring, these variations can result in unreliable identification of matching spectra from the spectral library for the measured spectrum.

0022To compensate for this, multiple libraries can be used with different libraries representing different patterns and different underlying layer thicknesses. A sequence of measured spectra can then be compared against reference spectra from multiple libraries to identify the library that gives the sequence the best fit, and the best fit library can be used to determine the polishing rate or end point. Can be used for.

0023In addition, substrates, especially device substrates, may have different zones with different characteristics, such as different feature densities or different underlying layer thicknesses. As a result, during spectroscopic monitoring performed in-situ during polishing, the spectra measured for one zone may not reliably match the reference spectra determined based on data from other zones.

0024This problem can be addressed by using multiple libraries representing different zones within the board. A sequence of current spectra of reflected light can be measured for each of multiple zones on a single substrate and for different zones to generate the best match reference spectrum that can be used to determine the endpoint. The spectra from the series can be compared against the reference spectra from different spectrum libraries.

0025Referring to FIG. 1, the substrate 10 is between the wafer 12, the outermost layer 14 to be polished, and the outermost layer 14 and the wafer 12, some of which are typically patterned1 It can include a layer or a plurality of underlying layers 16. For example, the outermost layer 14 and the immediately adjacent base layer 16 can both be dielectrics, for example, the outermost layer 14 can be an oxide film, and the immediately adjacent base layer 16 can be a nitride film. be able to. Other layers, such as other conductive layers and dielectric layers, can be formed between the immediately adjacent base layer and the substrate.

0026One potential problem with spectral end point detection during chemical mechanical polishing, especially when both the outermost layer 14 and the underlying layer 16 are dielectrics, is that of the underlying layer (s). The thickness (s) may vary between substrates or between zones on a single substrate. The substrate can have a plurality of zones such as a central zone, an intermediate zone, and an edge zone. For example, on a 300 mm wafer, the central zone can extend from a center to a radius of 50 mm, the intermediate zone can extend from a radius of 50 mm to about 100 mm, and the edges can extend from about 100 mm to about 150 mm. Can be done. In some implementations, the board has more or less zones than the three described.

0027As a result, multiple substrates (or multiple zones on a single substrate) with the same outermost layer may actually reflect different spectra depending on the underlying layer (s). .. As a result, the target spectrum used to trigger the polishing endpoint for one substrate (or zone of the substrate) is, for example, to another substrate (or zone of the substrate) if the underlying layers have different thicknesses. On the other hand, it may not function properly. However, when multiple spectra represent variations in the underlying layer (one or more layers), this effect can be compensated for by comparing the spectra obtained during polishing with the multiple spectra. ..

0028Changes in the thickness of the outermost layer to be polished at the start, changes in the thickness of the outermost layer during polishing (eg, due to different polishing rates in each zone), changes in the optical properties of the environment, below. A single substrate due to variations other than the thickness of the underlying layer, such as variations in the pattern of the formation, eg, line widths (eg, metal or polysilicon wire widths), or variations in layer composition. Fluctuations may still be inherently present between reference spectra determined using another substrate for or using another zone for one zone on the substrate. However, when multiple spectra represent other variations between substrates, it is similarly possible to compensate for this effect by comparing the spectra obtained during polishing with the spectra.

0029In addition, multiple libraries of reference spectra can be used to compensate for variability. What is in each library is provided with variations in the thickness of the outermost layer, but with multiple reference spectra representing substrates (or zones) that otherwise have similar characteristics, eg, similar underlying layer thickness. is there. Other variations, such as variations in the thickness of the underlying layer (s) between libraries, can be represented, for example, different libraries comprising different thicknesses of the underlying layer (s). Contains a reference spectrum representing the substrate (or zone) to be.

0030FIG. 2 is a cross-sectional view illustrating an example of a polishing apparatus 20 that operates so as to polish the substrate 10. The polishing apparatus 20 includes a rotatable disk-shaped platen 24 on which the polishing pad 30 is placed. The platen operates to rotate around the axis 25. For example, the motor can rotate the drive shaft 22 to rotate the platen 24.

0031Optical access 36 through the polishing pad is provided by including an aperture (ie, a hole through the pad) or a three-dimensional window. In one implementation example, a three-dimensional window can be supported on the platen 24 and protrude into the aperture in the polishing pad, but the three-dimensional window can be fixed to the polishing pad. The polishing pad 30 is usually placed on the platen 24 so that the aperture or window overlaps the optical head 53 placed in the recess 26 of the platen 24. The optical head 53 has, as a result, optical access to the substrate to be polished through the aperture or window. The optical head will be further described below.

0032The polishing apparatus 20 includes a composite slurry / rinse arm 39. During polishing, the arm 39 operates to administer a polishing liquid 38 such as a slurry. Alternatively, the polishing apparatus includes a slurry port that operates to administer the slurry onto the polishing pad 30.

0033The polishing apparatus 20 includes a carrier head 70 that operates to hold the substrate 10 against the polishing pad 30. The carrier head 70 is suspended from a support structure 72, for example a carousel, and is connected to the carrier head rotation motor 76 by a carrier drive shaft 74 so that the carrier head can rotate around a shaft 71. In addition, the carrier head 70 can reciprocate laterally through the radial slots formed in the support structure 72. In operation, the platen is rotated around its own central axis 25 and the carrier head is rotated around its own central axis 71, laterally moving over the entire upper surface of the polishing pad.

0034The polishing apparatus also includes an optical monitoring system that can be used to determine the polishing end point or whether to adjust the polishing rate (or adjust the polishing rate) as described below. The optical monitoring system includes a light source 51 and a photodetector 52. Light passes through the optical access 36 in the polishing pad 30 from the light source 51, hits the substrate 10, is reflected from the substrate 10 and returns through the optical access 36, and proceeds to the photodetector 52.

0035The bifurcated optical cable 54 can be used to transmit light from the light source 51 to the optical access 36 and light from the optical access 36 back to the photodetector 52. The bifurcated optical cable 54 can include a "main line" 55 and two "branch lines" 56 and 58.

0036As mentioned above, the platen 24 includes a recess 26 in which the optical head 53 is placed. The optical head 53 holds one end of the trunk line 55 of the bifurcated fiber cable 54, and the bifurcated fiber cable 54 is configured to transmit light to and from the surface of the substrate being polished. The optical head 53 may include one or more lenses or windows that overlap the ends of the bifurcated fiber cable 54. Alternatively, the optical head 53 can simply hold the end of the trunk line 55 adjacent to the three-dimensional window in the polishing pad.

0037The platen contains a removable in-situ monitoring module 50. The in-situ monitoring module 50 includes one or more of a light source 51, a photodetector 52, and a circuit for sending and receiving signals from the light source 51 and the photodetector 52. Can be done. For example, the output of the detector 52 may be a digital electronic signal, which passes through a rotary coupler in the drive shaft 22, eg, a slip ring, to a controller 60, such as a computer, for an optical monitoring system. Similarly, the light source can be turned on or off depending on the control command in the digital electronic signal passing through the rotary coupler from the controller to the module 50.

0038The in-situ monitoring module can also hold the respective ends of the branch line portions 56 and 58 of the bifurcated optical fiber 54. The light source operates to send light, which travels through the branch line 56, exits from the end of the trunk line 55 installed in the optical head 53, and hits the substrate being polished. The light reflected from the substrate is received at the end of the trunk line 55 installed in the optical head 53 and transmitted to the photodetector 52 through the branch line 58.

0039The light source 51 operates so as to emit white light. In one implementation example, the emitted white light comprises light having a wavelength of 200-800 nanometers. Suitable light sources are xenon lamps or xenon mercury lamps.

0040The photodetector 52 can be a spectroscope. A spectroscope is basically an optical instrument for measuring the intensity of light over a portion of an electromagnetic spectrum. A suitable spectroscope is a grating spectroscope. A typical output for a spectroscope is the intensity of light as a function of wavelength (or frequency).

0041The light source 51 and the photodetector 52 are connected to an arithmetic device that operates to control these operations and receive these signals, for example, a controller 60. The computing device can include a microprocessor installed near the polishing device, for example, a personal computer. In terms of control, the computing device can, for example, synchronize the activation of the light source 51 with the rotation of the platen 24.

0042As the platen rotates, as shown in FIG. 3, the computer can emit a series of flashes to the light source 51 that begins just before the substrate 10 passes over the in-situ monitoring module and ends shortly after (illustrated). Each of the points 301 to 311 represents the place where the light from the in-situ monitoring module is hit and reflected). Alternatively, the computer can emit a continuous light to the light source 51 that begins just before the substrate 10 passes over the in-situ monitoring module and ends shortly thereafter. In either case, the signal from the detector can be integrated over the sampling period to generate a spectral measurement at the sampling frequency. The sampling frequency can be about 3 to 100 milliseconds. Although not shown, each time the substrate 10 passes over the monitoring module, the alignment of the substrate with the monitoring module may be different from the previous pass. The spectra are obtained from different radii on the substrate over one revolution of the platen. That is, some spectra are obtained near the center of the substrate and some are near the edges. In addition, a series of spectra can be obtained over time over multiple rotations of the platen.

0043In operation, the computing device may receive a signal carrying information describing the spectrum of light received by the photodetector 52, for example, during a particular flash of light or during the time frame of the detector. it can. Therefore, this spectrum is a spectrum measured in-situ during polishing.

0044Without being limited to any particular theory, the spectrum of light reflected from substrate 10 changes progressively as polishing progresses due to changes in the thickness of the outermost layer, and thus of a time-varying spectrum. Create a series. Moreover, there is a distinctive spectrum due to the unique thickness of the stacked layers.

0045The computing device can process the signal to determine the end point of the polishing step. In particular, the computing device can carry out the logic of determining when the end point has been reached, based on the measured spectrum.

0046Briefly, the computing device can compare the measured spectra against multiple reference spectra and use the results of the comparison to determine when the end point has been reached.

0047As used herein, the reference spectrum is a pre-determined spectrum prior to polishing the substrate. The reference spectrum can have a predetermined relationship with the value of the substrate property, such as the thickness of the outermost layer, that is, a predetermined relationship prior to the polishing operation. As an alternative or in addition, assuming that the actual polishing rate follows the expected polishing rate, the reference spectrum is predetermined with a value representing the time during the polishing process that the spectrum is expected to appear at that time. Can have a relationship.

0048The reference spectrum can be generated empirically or from theory, for example by measuring the spectrum from a test substrate with a known layer thickness. For example, in order to determine the reference spectrum, the spectrum of a "setting" substrate having the same pattern as the product substrate can be measured at the measurement station before polishing. Substrate properties, such as the thickness of the outermost layer, can also be measured prior to polishing using the same or different measuring stations. The setting substrate is then polished while collecting spectra. For each spectrum, record a value that represents the time during the polishing process that the spectrum was collected at that time. For example, the value can be the elapsed time or the number of revolutions of the platen. The substrate can be over-polished, i.e., polished past the desired thickness so that the spectrum of light reflected from the substrate can be obtained when the target thickness is achieved. The spectrum and properties of the setting substrate, eg, the thickness of the outermost layer, can then be measured at the measuring station after polishing.

0049Optionally, the set substrate can be removed from the polishing system on a regular basis and the properties and / or spectrum of the set substrate can be measured at the measuring station before returning to polishing. A value can also be recorded in which the spectrum represents the time during the polishing process measured at the measuring station at that time.

0050The reference spectrum is stored in the library. Reference spectra in the library represent substrates with varying thicknesses on the outer layer.

0051Multiple libraries can be created from different setting substrates that differ in characteristics other than the thickness of the outermost layer, for example, the thickness of the underlying layer, the pattern of the underlying layer, or the composition of the outer layer or the underlying layer.

0052From the collected spectra, the measured thickness and the collected spectra are selected to select one or more spectra that are determined to be indicated by the substrate when the substrate has a thickness of interest. use. In particular, linear interpolation can be performed using the measured pre-polishing film thickness and post-polishing substrate thickness to determine the time when the target thickness is achieved and the corresponding spectrum shown. One spectrum or multiple spectra that are determined to be shown when the target thickness is achieved are designated as one target spectrum or multiple target spectra.

0053In addition, the measured pre-polishing film thickness and post-polishing substrate thickness (or measured at a measuring station) were measured based on the time the spectra were collected and the time input of the measured spectra, assuming a uniform polishing rate. Linear interpolation with (other thicknesses) can be used to calculate the thickness of the outermost layer for each spectrum collected in-situ. Due to the initial flattening, the polishing rate may not be uniform from the beginning to the end of the polishing operation, in this case the thickness is still calculated if a gradual change in the polishing rate is known. can do. In addition, it can be assumed that the speed can be uniform towards the end of polishing.

0054In addition to empirically determining, some or all of the reference spectra can be calculated from theory, for example using an optical model of the substrate layer. For example, an optical model can be used to calculate the spectrum for a given outer layer thickness D. A value representing the time during the polishing process in which the spectrum should be collected at that time can be calculated, for example, by assuming that the outer layer is removed at a uniform polishing rate. For example, the time Ts for a particular spectrum can be easily calculated by assuming a starting thickness D0 and a uniform polishing rate R (Ts = (D0-D) / R). As another example, the measurement time T for the pre-polishing thickness D1 and the post-polishing thickness D2 (or any other thickness measured at the measuring station) based on the thickness D used for the optical model.₁, T₂Can perform linear interpolation between (Ts = T)₂-T₁* (D1-D) / (D1-D2)).

0055As used herein, a library of reference spectra is a collection of reference spectra representing substrates that share common properties (other than the thickness of the outer layer). However, properties that are commonly shared within a library may change across multiple libraries in the reference spectrum. For example, two different libraries can contain reference spectra representing substrates with two different substrate thicknesses.

0056Spectrums for different libraries, by polishing multiple "setting" substrates with different substrate properties (eg, underlayer thickness, or layer composition), and by collecting spectra as discussed above. Can be generated; spectra from one set substrate can give a first library and spectra from another substrate with different underlying layer thickness can give a second library. As an alternative or in addition, reference spectra for different libraries can be calculated from theory, for example, spectra for the first library using an optical model with a first thickness underlayer. And the spectrum for the second library can be calculated using an optical model with underlying layers with different thicknesses.

0057In one implementation example, each reference spectrum is assigned an index value. This index can be a value representing the time during the polishing process where the reference spectrum is expected to be observed at that time. The spectra can be indexed so that each spectrum in a particular library has a unique index value. The spectra can be indexed so that the index values are ordered in the order in which each spectrum is measured. The index value can be selected to change monotonically as polishing progresses, for example to increase or decrease. In particular, the index value of the reference spectrum can be selected so that the index value forms a linear function of time or platen speed. For example, the index value can be proportional to the platen speed. Thus, each index number can be an integer, which can represent the expected platen rotation in which the relevant spectrum should appear at that rotation.

0058The reference spectrum and related indexes can be stored in the library. The library can be implemented in the memory of the computing device of the polishing machine. The index of the target spectrum can be called the target index.

0059During polishing, index marks can be generated for each library. Each index trace contains a series of indexes that form the trace, and each particular index in the series is associated with a particular measured spectrum. For the index traces of a given library, a particular index in the series is generated by selecting the index of the reference spectrum from the given library that best fits the particular measured spectrum.

0060As shown in FIG. 4, the index 80 corresponding to each measured spectrum can be plotted against time or platen rotation. A known degree polynomial function, eg, a linear function (ie, a straight line), is fitted to the plotted index number, eg, using robust straight line fitting. The intersection of the straight line with the target index determines the end time or end rotation. For example, the linear function 82 is fitted to the data points as shown in FIG.

0061Without being limited to any particular theory, one library can more accurately predict reasonable endpoints than others because it matches the data measured more consistently by one library. For example, among multiple libraries representing substrates (or zones within a substrate) with different substrate thicknesses, it most closely matches the measured substrate thickness (or zones within that substrate). The library should give you the best match. Therefore, the advantage of the present invention is a more accurate end point detection system realized by utilizing a plurality of reference spectrum libraries. In particular, different reference spectrum libraries can be used for each zone of the substrate. In addition, each zone can have a number of different reference spectrum libraries.

0062For example, FIG. 4 is a schematic diagram illustrating an example index trace from a spectroscopic monitoring system showing good data fit corresponding to a first zone on a substrate. In comparison, FIG. 5 is a schematic diagram illustrating an example index trace from a spectroscopic monitoring system showing poor data fit to a second zone on its substrate. The index traces in the examples in Figures 4 and 5 represent index traces generated using the same reference spectrum library. Compared to the difference between the robust straight line related to the number of indexes plotted in FIG. 4 and the related index trace, the number of indexes plotted in FIG. 5 is relatively large from the relevant robust straight line. Have a difference. Therefore, it may be advantageous to use different reference spectrum libraries for different zones on the substrate.

0063In one implementation, the spectrum is obtained at more than one radial position on the board. For each spectral measurement, a radial position on the substrate can be determined and the spectral measurements can be grouped into zones based on their own radial position (eg, radial zones). The substrate can have a plurality of zones, such as a central zone, an intermediate zone, and an edge zone, as described above. By using the method described in US Pat. No. 7,097,537, or as described in US Pat. No. 7,018,271, which is incorporated herein by reference for all purposes, etc. , The position where the spectrum is obtained from it can be determined.

0064The spectra measured from each zone (or, for each zone, the average of the spectra from within the zones obtained from a single sweep of the sensor across the substrate), are from multiple reference spectrum libraries, as described above. The reference spectrum in one or more of them is compared and the corresponding index number is determined by comparison with the spectrum library. The corresponding number of indexes for each zone can be used to generate the index traces, and the index traces can be used to determine the goodness of fit.

0065FIG. 6 shows method 600 for determining the end point of a polishing step. Polish one substrate from a batch of substrates (step 602) and perform the following steps for each platen rotation (and for each zone of the substrate where multiple libraries are used for each zone). One or more spectra are measured to obtain the current spectrum for the current platen rotational motion (step 604). Determine the first best match reference spectrum stored in the first spectrum library that best fits the current spectrum (step 606). Determine the second best match reference spectrum stored in the second spectrum library that best fits the current spectrum (step 608). More generally, for each library you are going to use for the substrate and / or zone, determine the reference spectrum that best matches the current spectrum. Determine the index of the first best match reference spectrum from the first library that best fits the current spectrum (step 610) and add it to the first index trace associated with the first library (step 612). ). Determine the index of the second best match reference spectrum from the second library that best fits the current spectrum (step 614) and add it to the second index trace associated with the second library (step 616). ). More generally, for each library, an index is determined for each best match reference spectrum and added to the index trace for the relevant library. The first straight line is fitted to the first index mark (step 620) and the second straight line is fitted to the second index mark (step 622). More generally, for each index mark, a straight line can be fitted to the index mark. Robust straight line fitting can be used to fit straight lines.

0066The index of the first best match spectrum matches or exceeds the target index (step 624), and the index trace associated with the first spectrum library has the best fit to the robust straight line associated with the first spectrum library (step 624). Step 626) When, or when the index of the second best match spectrum matches or exceeds the target index (step 624), the index trace associated with the second spectral library goes to the robust straight line associated with the second spectral library. The end point can be declared when it has the best fit of (step 626) (step 630). More generally, the end point can be declared when the index trace, which is the best fit for the relevant fitting line, matches or exceeds the target index.

0067Although the two libraries are discussed above, the technology can be used with three or more libraries. In addition, some or all of the libraries may or may not be shared between zones, for example, some or all of the libraries for one zone may be shared in another zone. May or may not be available.

0068Also, rather than comparing the index value itself to the target index, the value of the straight line fitted at the current time can be compared to the target index. That is, a value (which does not necessarily have to be an integer in this situation) is calculated from the linear function for the current time and this value is compared against the target index.

0069Method 600 can be used, for example, to use different reference spectrum libraries to determine polishing endpoints for different zones of the substrate. In particular, a reference spectrum library is used, which provides an index trace with the best goodness of fit for a particular zone. In these or other implementations, some zones may use the same reference spectrum library, but some zones may use different reference spectrum libraries. In one implementation, a subset of multiple reference spectrum libraries can be pre-determined (eg, selected by the user) to limit the number of libraries used for each zone. For example, two or more reference libraries can be predetermined for use in each zone. In one implementation example, a particular reference spectrum library can be identified for each zone based on goodness of fit. For example, during a given time interval during the polishing process (eg, the first 10-20 seconds of polishing), a particular reference spectrum library to be used for each zone, eg, the best library for one zone, is defined. Can be selected based on a reference spectrum library that provides the best goodness of fit during the time interval of.

0070Other implementation examples are possible. For example, although two libraries are discussed above, the technology can be used with three or more libraries. As another example, some or all of the libraries may or may not be shared between zones, for example, some or all libraries for one zone in another zone. May or may not be available for. As yet another example, just one reference spectrum library can be pre-determined for use in each zone, just as different reference spectrum libraries are used for each zone. For this implementation example, the reliability of pointing to the end point can be improved by not selecting reference libraries based on goodness of fit, but rather by simply using different reference libraries for different zones.

0071Determining whether an index trace associated with a spectral library has the best fit for a linear function associated with a library is for differences from the associated robust lines and index traces associated with another library. By comparison, the index traces associated with the spectral library have the smallest magnitude difference from the associated robust line, eg, the smallest standard deviation, the largest correlation, or other measure of variation. Can include determining whether or not to have. In one implementation, the goodness of fit is determined by calculating the sum of squares of the differences between the index data points and the linear function, and the library with the least sum of squares of the differences has the best fit.

0072If one of the index traces reaches the target index but is not the best fit, the system either until the index trace reaches the best fit or until the best fit index trace reaches the target index. Can wait.

0073Although only two libraries and two index traces are discussed above, the concept is applicable to two or more libraries that should give more than one index trace. In addition, rather than declaring the end point when the trace index matches the target index, it is possible to declare the end point when calculating the linear fitting to the trace so that it intersects the target index. Moreover, in order to reduce processing, it is possible to eliminate the worst fit index marks, for example, from about 40% to 50% or 60% of the total expected polishing time, before the end point. There should be.

0074Obtaining the current spectrum can include measuring at least one spectrum of light reflected from the surface of the substrate being polished (step 604). Multiple spectra can be optionally measured, eg, spectra measured at different radii on the substrate can be obtained from one rotation of the platen, eg, at points 301-311 (FIG. 3). it can. When measuring multiple spectra, one or a subset of the spectra can be selected for use in the endpoint detection algorithm. For example, spectra measured at sample locations near the center of the substrate (eg, at points 305, 306, and 307 shown in FIG. 3) can be selected. To increase accuracy and / or accuracy, the spectra measured during the current platen rotational motion are optionally processed.

0075Determining the difference between each of the selected measured spectra and each of the reference spectra (step 606 or 610) can include calculating the difference as the sum of the intensity differences over a wavelength range. .. That is,

0076<img id="000002" he="18" wi="154" file="JP5774482B2_D0001.tif" img-format="tif" img-content="drawing" />Where a and b are the lower and upper limits of the wavelength range of the spectrum, respectively, and I_current(λ) and I_reference(λ) is the intensity of the current spectrum and the intensity of the reference spectrum for a given wavelength, respectively. Alternatively, the difference can be calculated as a mean square error, i.e.

0077<img id="000003" he="21" wi="153" file="JP5774482B2_D0001.tif" img-format="tif" img-content="drawing" />

0078If there are multiple current spectra from a given platen rotation for the substrate or for each zone of the substrate, the best match is determined between each of the current spectra and each of the reference spectra of a given library. that you can. Each selected current spectrum is compared against each reference spectrum. Given the given current spectra e, f and g and the reference spectra E, F and G, for example, the matching coefficients are set to the following combinations of the current and reference spectra: e and E, e and F, respectively. , E and G, f and E, f and F, f and G, g and E, g and F, and g and G: can be calculated. Which matching factor indicates the best match, for example the smallest, determines the reference spectrum and thus the index.

0079Determining whether the index traces associated with the spectral library have the best fit for the robust straight lines associated with the spectral library (step 620 or 624) relates to the data points and spectral libraries that contain the index traces. The minimum sum of squares of the difference between the library and the fitted Robust line, eg, the minimum difference between the data points as shown in FIGS. 4 and 5 and the associated Robust line. Includes determining whether to have a sum of squares.

0080In one implementation example, the expected end time is determined for one zone, such as the central zone. Where appropriate, the polishing rate in the other zones is then adjusted to achieve the desired end point at the same time as the expected end time for the selected zone, eg, the central zone. The polishing rate can be adjusted, such as by increasing or decreasing the pressure in the corresponding zone within the carrier head. In some carrier heads, such as those described in US Patent Application Publication No. 2005/0211377, the carrier head has an adjustable pressure zone. For example, a simple Preston model can be assumed such that the change in polishing rate is directly proportional to the change in pressure. In addition, controls for polishing the substrate, taking into account platen or head rotation speed, secondary effects of different head pressure combinations, polishing temperature, slurry flow, or other parameters that affect polishing speed. Models can be developed.

0081With reference to FIG. 7, if a particular profile is desired, such as a uniform thickness over the entire surface of the substrate, the gradient of polishing rate as indicated by the change in the number of indexes over time is monitored. If the goodness of fit of the index trace indicates that the spectral measurements are reliable (eg, the goodness of fit is less than a predetermined threshold), the polishing rate can be adjusted. After the polishing stabilization period 705, the spectra are obtained at the central zone 710, at the edge zone 715, and at the intermediate zone 720 in between. Here, the zone is a circular zone or a circular zone. Associate each spectrum with its own index. This process is repeated over platen speeds or over time to determine the polishing rate in each of the central zone 710, intermediate zone 720, and edge zone 715. The rate of polishing is dictated by the gradient of the straight line obtained by plotting the index 730 (y-axis) according to the number of revolutions 735 (x-axis). If any of the velocities are calculated to be faster or slower than the others, then the velocity in the zone can be adjusted if the goodness of fit of the index trace indicates that the spectral measurements are reliable. it can. Here, the adjustment is the end point C of the central zone 710._Ebased on. For some implementations, no adjustment is required if the polishing rate is within acceptable tolerances. Approximate polishing end points EDPs are known by polishing similar substrates with similar polishing parameters or by using the different methods described above. First polishing time T during the polishing process₁In, the rate of polishing at the intermediate zone 720 is reduced and the rate of polishing at the edge zone is increased. If you do not adjust the polishing rate at the intermediate zone 720, the intermediate zone should be polished faster than the rest of the substrate, M._AIt is polished at the over-polishing speed of. About end zone 715 T₁If the polishing rate is not adjusted in, the edge zone 715 is E_uShould be under-polished at the speed of.

0082Subsequent time during the polishing process (T₂), The speed can be adjusted again if appropriate. The goal of this polishing process is to finish polishing when the substrate has a flat surface, or an oxide layer over the entire surface that is relatively smooth. One way to determine the amount to adjust the rate of polishing is to adjust the rate so that the indexes of the central zone, intermediate zone, and edge zone are equal at the approximate polishing end point EDP. Therefore, T the central zone and the middle zone₂It is necessary to adjust the polishing rate at the edge zone while polishing at the same rate as before. If the EDP is approximate, polishing can be stopped when the index in each zone is at the desired location, i.e. when each location has the same index.

0083It is preferable to change the polishing rate several times during the polishing process, such as four times, three times, two times, or only one time. Adjustments can be made near the beginning of the polishing process, in the middle, or towards the end. Correlating the spectrum to the number of indexes can create a linear comparison at each part of the zone and simplify the calculations required to determine how to control the polishing process, complex software. Alternatively, the processing step can be removed.

0084A method that can be applied during the endpoint process is to limit the portion of the library searched for the matching spectrum. The library contains a wider spectrum than what would typically be obtained while polishing the substrate. The wide range describes the spectrum obtained from the thicker starting outermost layer and the spectrum obtained after overpolishing. During substrate polishing, the library search is limited to a predetermined range of the library spectrum. In one embodiment, the current rotational index N of the substrate being polished is determined. N can be determined by searching the entire library spectrum. Search the library within N degrees of freedom for the spectra obtained during subsequent rotations. That is, if the number of indexes is found to be N during one rotation, the range is from (N + X) -Y to (N) during subsequent rotations where the degrees of freedom are Y and after X rotations. Search up to + X) + Y. For example, if the matching index is found to be 8 and the degrees of freedom are selected to be 5 on the first polishing rotation of the substrate, then the index with respect to the spectrum obtained during the second rotation. Only the spectra corresponding to the number 9 ± 5 are investigated for matching. When applying this method, the same method can be applied independently to all of the libraries currently used in the end point detection process.

0085All of the embodiments of the present invention and the functional operations described herein are in digital electronic circuits, computer software, firmware, or the structural means disclosed herein and their structural equivalents. It can be implemented in hardware containing things, or in a combination of these. Embodiments of the invention as one or more computer program products, i.e., for execution by a data processing device, eg, a programmable processor, a computer, or a plurality of computers or computers, or a data processing device. Can be implemented in an information carrier, eg, in a machine-readable storage medium, as one or more computer programs tangibly embodied to control the behavior of. A stand-alone program that allows you to write a computer program (also known as a program, software, software application, or code) in any form of programming language, including a compiled or interpreted language. It can be placed in any format, including as a module, as a component, or as another unit suitable for use in a computer environment. Computer programs do not necessarily have to deal with files. Place the program in a part of a file that holds other programs or data, in a file dedicated to the program in question, or in multiple organized files (eg, one or more modules). It can be stored in a subprogram, or a file that stores part of the code). Computer programs can be arranged to run on one computer or on multiple computers distributed across one site or sites and interconnected by communication networks.

0086One or more programmable processes and logic flows described herein that perform one or more computer programs to run on input data and perform functions by producing output. Can be run by any processor. Process and logic flows can also be carried out by special purpose logic circuits, such as FPGAs (Field Programmable Gate Arrays) or ASICs (Application Specific Integrated Circuits), and the equipment can also be implemented as the circuits described above. Can be done.

0087The polishing equipment and methods described above can be applied to various polishing systems. Either or both of the polishing pad and the carrier head can be moved to provide relative movement between the polishing surface and the substrate. For example, a platen can revolve rather than rotate. The polishing pad can be a circular (or some other shape) pad fixed to the platen. Some aspects of end point detection systems may be applicable, for example, to linear polishing systems in which linearly moving polishing pads are continuous belts or reel-to-reel belts. The polishing layer can be a standard (eg, polyurethane with or without filler) polishing material, a soft material, or a fixed abrasive grain material. It should be understood that the polished surface and substrate can be held vertically or in one other direction using the term relative positioning.

0088Specific embodiments of the present invention have been described. Other embodiments are within the scope of the claims below. For example, the actions described in the claims can be performed in a different order and the desired result can still be achieved.