JP4563811B2 - Interferometry and scanning interferometers for ellipsometry, reflected light and scattered light measurements, including characterization of thin film structures - Google Patents

Description

  The present invention relates to surface topography measurement of an object having a hybrid structure made of a thin film or different materials. Such measurements relate to characterization of flat panel display components, semiconductor wafer measurements, and in situ analysis of thin films and dissimilar materials.

  Ellipsometry is used to analyze the optical properties of complex surfaces. Ellipsometry is based on the difference in complex reflectivity of the surface when illuminated at a certain tilt angle, eg 60 °, but variable angles or multiple wavelengths may be used. Numerous types of ellipsometers are known in the art.

  In order to achieve a higher resolution than can easily be achieved in a conventional ellipsometer, the microscopic ellipsometer measures the phase and / or intensity distribution in the back focal plane of the object. This is also known as the pupil plane, and various illumination angles are mapped to the field position. Such devices are modern in style with traditional polarizing microscopes or “conoscopes” that analyze the pupil plane of birefringent materials using crossed polarizers and Bertrand lenses, historically related to crystallography and mineralogy. Is modified.

  Embodiments of the present invention, at least in part, allow different angles of incidence of an interferometer (eg, having a high NA objective lens) to scan a test sample or reference mirror (eg, test) against the interferometer. The interference fringes generated by moving up and down relative to the objective lens used to focus the light on the sample or reference mirror are based on a method that can be distinguished by the corresponding spatial frequency. Therefore, by mathematically spatial frequency resolving such interference fringes, the relative amplitude and phase of light reflected (or scattered) from the sample surface can be determined as a function of angle. The information obtained by this, together with the calibration of the illumination distribution of the pupil plane of the objective lens and the polarization state of the illumination across the pupil plane, does not require the pupil plane to be imaged directly on the detector array, Multi-angle reflection (or scattering) amplitude and phase information is provided for each pixel in the field of view. Using these multi-angle data, sample surface properties such as thin film thickness and / or complex refractive index can be obtained with high lateral resolution for each pixel, as well as surface height profile information.

Embodiments of the present invention typically include an interferometer, for example an interference microscope comprising an interference objective such as a Mirau, Linnik, Michelson or the like. The objective lens illuminates the sample surface and collects light from that surface over a range of incident angles φ. For example, in the case of an interference objective lens having a numerical aperture (NA) of about 0.75, φ = 0 to 50 °. As illumination polarization, radial polarization, linear polarization, circular polarization, polarization depending on the field of view, or adjustable polarization can be used. Typically, the apparatus further samples the sample along an axis parallel to the optical axis of the objective lens while collecting interference intensity data for an array of pixels corresponding to the field position on the sample with an electronic camera. A mechanical scanner is also provided for moving the surface (or moving the objective lens relative to the sample, correspondingly). Alternatively, the reference leg of the interferometer
Can also be scanned. As a result, intensity versus sample position data is stored in computer memory for each pixel for a series of objective lens distances from the sample.

  In some embodiments, the computer converts the per-pixel interference data to the frequency domain, for example by Fourier analysis, to recover the magnitude and phase of the spatial frequency components present in the interference data. The computer analyzes this data and compares its magnitude and phase to a model that represents the surface structure of the sample, including optical properties that depend on the angle of incidence, polarization, and / or wavelength of the sample. This analysis identifies parameters such as surface height or thin film thickness.

  In some embodiments, in addition to selecting the wavelength to the interferometer or sending multiple wavelengths to the interferometer to analyze the angular dependence of the material's optical properties, the optical properties of the material as a function of wavelength Perform detailed analysis. In some embodiments, light scattered from the sample is analyzed to identify surface structure information by surface diffraction and scattering characteristics as a function of incident angle and wavelength.

  Embodiments of the present invention include a number of advantages. For example, these embodiments resolve the fringes generated by scanning the sample vertically with respect to the interfering objective lens in the frequency domain, e.g., for each pixel, its optical properties and surface topography. Means can be provided for analyzing the surface structure to obtain simultaneously. Such an approach uses both amplitude and phase information from the reflected light without directly using the instrument's pupil plane, and optical properties and wavelength dependent optics that depend on the angle of the surface. Properties can be obtained.

Here, various aspects and features of one or more embodiments of the present invention are summarized in a comprehensive manner.
In general, in one aspect, a feature of the invention is to image test light emanating from a test object over a range of angles so as to interfere with reference light on a detector, The reference light is generated from a common light source. Simultaneously changing the optical path length difference from the light source to the detector between the interfering portions of the test light and the reference light at each angle at a speed depending on the angle at which the test light exits the test object; and While changing the optical path length difference for each angle, the angle dependency of the optical characteristics of the test object is specified based on the interference between the test light and the reference light.

Embodiments of the method can include any of the following features.
The range of incident angles may correspond to a numerical aperture greater than 0.7, or more preferably greater than 0.9.

  The detector may be a camera having a plurality of detector elements, and imaging consists of imaging test light emanating from various locations of the test object at corresponding locations on the camera. Further, identifying the angular dependence of the optical properties includes identifying the angular dependence of the optical properties for various locations of the test object.

  The angular dependence of the optical property may be related to a change in the optical property as a function of the angle of the test light incident on the test object. The method may further include illuminating the plurality of locations on the test object with the test light such that the test light is incident on each of the plurality of locations over a range of incident angles. In such a case, illuminating and imaging may include using a common objective lens. Furthermore, the common light source may be a spatially extended light source.

In other embodiments, the angular dependence of the optical property is related to a change in the optical property as a function of the angle of the test light scattered (ie, diffracted) from the test object. The method further includes illuminating multiple locations of the test object with test light having a uniform angle of incidence on the test object, and the imaging is scattered from each location of the test object over a range of angles. Imaging the test light at a corresponding location on the detector may be included. In such cases, illumination and imaging may include using a common objective lens. Further, the common light source can be a point light source.

Imaging may further include polarizing test light at a pupil plane of an optical system involved in imaging.
The method may further include illuminating the test object with the test light and polarizing the test light at a pupil plane of the optical system used to illuminate the test object.

The common light source can be monochromatic. For example, a common light source may have a central wavelength and a spectral bandwidth that is less than 2% of the central wavelength.
Simultaneously changing the optical path length difference for each angle may include moving the test object relative to the objective lens used to collect the test light emerging from the test sample.

  Simultaneously changing the optical path length difference for each angle may include moving a reference mirror used to reflect the reference light relative to an objective lens used to focus the reference light on the reference mirror.

Simultaneously changing the optical path length difference for each angle may include moving a beam splitter disposed within the mirro interference objective lens.
Changing the optical path length difference for each angle simultaneously may define the spatial coherence length, and the optical path length difference for at least one of the angles may be changed over a range greater than the spatial coherence length.

  Identifying the angular dependence of the optical properties is to measure the interference signal from the detector as the optical path length difference is changed simultaneously for each angle, and linear to the optical path length difference for each angle. Transforming the interference signal to a coordinate proportional to, thereby generating and transforming a transformed signal that depends on a conjugate variable to the coordinate. For example, the conjugate variable can be a spatial frequency.

  The conjugate variable can provide a direct mapping to the angle of test light incident on or emitted from the test object. For example, if the conjugate variable is spatial frequency K, a direct mapping between the spatial frequency and the angle φ can be given by K (φ) ∝cos (φ) / λ, where λ is the wavelength of the test light is there. For example, when the emitted light is reflected from the test sample, a direct mapping between spatial frequency and angle is given by K (φ) = 4πcos (φ) / λ.

The transformed signal can give a direct mapping to the angular dependence of the optical properties. For example, the transform may correspond to a Fourier transform.
The optical property is related to the complex reflectivity of the test object. For example, the optical properties are related to the magnitude of the complex reflectivity of the test object. The optical characteristics are also related to the phase of the complex reflectivity of the test object.

  The angular dependence of the optical characteristics is a characteristic that depends on the interference between the test light and the reference light when the optical path length difference is changed for each angle, and the pre-calibrated angle of the optical system related to imaging. Can be identified based on

The method may further include identifying a surface height profile of the test object based on interference between the test light and the reference light when the optical path length difference is changed.
The method may further include comparing the change in angular dependence of the optical properties identified from the interference between the test light and the reference light with the change in angular dependence of the model for the test object. For example, the test object may include at least one thin film on the substrate, and the method may further include determining the thickness of the thin film based on the comparison.

  In one such embodiment, the optical characteristics include the magnitude of the angular dependence of the complex reflectivity of the test sample, and the thickness of the thin film can be determined by measuring the magnitude of the angular dependence of the complex reflectance and the angular dependence of the model. Based on comparing size. Further, the present embodiment may include identifying a surface height profile for the test object based on the comparison. For example, the optical properties may further include the angle-dependent phase of the complex reflectivity of the test sample, and the identification of the surface height profile can be determined by specifying the specified thickness of the thin film and the angle-dependent phase of the complex reflectivity. Based on comparing the angle-dependent phase of the model for a given thickness.

  Finally, the test light and reference light may have a first wavelength, and the method includes imaging in the case of test light and reference light having a second wavelength different from the first wavelength; It may further comprise repeating changing and identifying.

  In general, in another aspect, the invention features a method that includes determining angular dependence of optical properties of a test object based on scanning interferometry data for the test object. And

The method may further include any of the features described above in connection with the first method.
In general, in yet another aspect, the invention is to image test light emanating from a test object over a range of angles so as to interfere with reference light on a detector, the test light And the reference light is generated from a common monochromatic light source, and the test object includes at least one thin film on the substrate. For each angle, simultaneously changing the optical path length difference from the light source to the detector between the interfering portions of the test light and the reference light at a speed depending on the angle at which the test light exits the test object, and the angle The method includes determining the thickness of the thin film based on the interference between the test light and the reference light while changing the optical path length difference every time.

  In general, in yet another aspect, the present invention identifies the thickness of a thin film on a test object based on monochromatic light scanning interferometry data for a test object that includes the thin film and a substrate that supports the thin film. A method comprising:

The third and fourth method embodiments described above may further include any of the features described above in connection with the first method.
In general, in yet another aspect, the present invention images a test light emanating from a test object over a range of angles so as to interfere with a light source, a detector, and a reference light on the detector. A scanning interferometer configured such that the test light and the reference light are generated from a light source, the scanning interferometer further comprising a test light and a reference at a rate depending on the angle at which the test light exits from the test object A scanning interferometer configured to simultaneously change the optical path length difference from the light source to the detector between the interfering portions of the light, and an electronic processor coupled to the detector and the scanning interferometer Configured to determine the angular dependence of the optical properties of the test object based on the interference between the test light and the reference light when the optical path length difference is changed from angle to angle as measured by the detector Electronic pro It features an apparatus and a Tsu service.

In general, in yet another aspect, the invention images a test light emitted from a test object over a range of angles so as to interfere with a monochromatic light source, a detector, and a reference light on the detector. A scanning interferometer configured so that the test light and the reference light are generated from a monochromatic light source, and for each angle, the scanning interferometer further depends on the angle at which the test light exits from the test object A scanning interferometer configured to simultaneously change the optical path length difference from the light source to the detector between the interfering portions of the test light and the reference light, and coupled to the detector and the scanning interferometer An electronic processor configured to determine a thickness of the thin film on the test object based on interference between the test light and the reference light when the optical path length difference is changed for each angle. A device comprising a processor And features.

  In general, in yet another aspect, the present invention relates to a scanning interferometry system and an electronic processor coupled to the scanning interferometry system for a test object generated by the scanning interferometry system. An apparatus comprising an electronic processor configured to identify angular dependence of optical properties of a test object based on scanning interferometry data.

  In general, in yet another aspect, the invention relates to a monochromatic optical scanning interferometry system and an electronic processor coupled to the scanning interferometry system, the monochromatic optical scanning interferometry data for a test object. And an electronic processor configured to determine the thickness of the thin film on the test object.

  In general, in yet another aspect, the present invention is a scanning system configured to image test light emanating from a test object over a range of angles so as to interfere with reference light on a detector. An apparatus comprising an interferometer, wherein the test light and the reference light are generated from a common light source, and for each angle, the scanning interferometer further comprises a test light at a speed depending on the angle at which the test light exits from the test object. And an interferometer arranged to collect test light emanating from the test object, the optical path length difference from the light source to the detector between the interfering portion of the reference light and the detector is changed simultaneously And at least one polarization optical system disposed on the pupil plane of the objective lens.

For example, at least one polarizing optical system may provide polarization that varies across the pupil plane.
The at least one polarization optical system can include a polarizer and at least one wave plate. For example, the at least one polarization optical system may include two wave plates arranged at different positions on the pupil plane.

  In general, in yet another aspect, the present invention is a scanning system configured to image test light emanating from a test object over a range of angles so as to interfere with reference light on a detector. An apparatus comprising an interferometer, wherein the test light and the reference light are generated from a common light source, and for each angle, the scanning interferometer further comprises a test light at a speed depending on the angle at which the test light exits from the test object. And the interferometer is configured to illuminate the test object with generally collimated light, wherein the interferometer is configured to simultaneously change the optical path length difference from the light source to the detector between the interfering portions of the reference light Features an apparatus comprising a module. For example, the apparatus may further include a common light source, and the common light source may be a monochromatic light source.

  Furthermore, any of the apparatus embodiments of the present invention described thus far can include any of the corresponding features described above in connection with the first method. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. All publications, patent applications, patents, and other cited references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control.

Other features, objects and advantages of the invention will be apparent from the detailed description set forth below.
Like reference numbers in the various drawings indicate common elements.

  FIG. 1 shows a Linnik-type scanning interferometer. Illumination light 102 from a light source (not shown) is partially transmitted by the beam splitter 104 to define the reference light 106 and partially reflected by the beam splitter 104 to define the measurement light 108. The measurement light is focused by a measurement objective lens 110 onto a test sample 112 (for example, a sample including a single layer thin film made of one material or a multilayer thin film made of a plurality of different materials). Similarly, the reference light is focused on the reference mirror 116 by the reference objective lens 114. The measurement and reference objectives preferably have common optical properties (eg, the same numerical aperture). Measurement light reflected (or scattered or diffracted) from the test sample 112 is returned through the measurement objective 110, transmitted by the beam splitter 104, and imaged on the detector 120 by the imaging lens 118. The Similarly, the reference light reflected from the reference mirror 116 is also returned through the reference objective lens 114, reflected by the beam splitter 104, imaged on the detector 120 by the imaging lens 118, and interferes with the measurement light. .

  For simplicity, FIG. 1 shows measurement light and reference light that are focused at specific points on the test sample and reference mirror, respectively, and then interfere at corresponding points on the detector. Such light corresponds to the portion of the illumination light that propagates perpendicular to the pupil plane for the measurement and reference paths of the interferometer. The other part of the illumination light eventually illuminates the test sample and other points on the reference mirror and is then imaged at the corresponding point on the detector. In FIG. 1, this is indicated by the dashed line 122, which corresponds to the chief ray emanating from various points on the test sample and imaged to corresponding points on the detector. The chief ray traverses the center of the pupil plane 124 of the measurement path, which is the rear focal plane of the measurement objective lens 110. Light exiting the test sample at an angle different from the chief ray angle traverses different locations on the pupil plane 124.

  In a preferred embodiment, the detector 120 is a multi-component (ie, multi-pixel) camera that individually detects interference between the measurement light and the reference light corresponding to various points on the test sample and the reference mirror. Measure (ie, give spatial resolution for interference fringes).

  A scanning stage 126 coupled to the test sample 112 scans the position of the test sample relative to the measurement objective 110 as indicated by the scan coordinate ζ in FIG. For example, the scanning stage can be based on a piezoelectric transducer (PZT). When the relative position of the test sample is being scanned, the detector 120 measures the intensity of optical interference at one or more pixels of the detector and sends it to the computer 128 for analysis.

  Since scanning is performed in an area where the measurement light is focused on the test sample, the scan is detected from the light source separately according to the angle of the measurement light incident on the test sample and the measurement light exiting from the test sample. Change the optical path length of the measurement light to the instrument. As a result, the optical path difference (OPD) from the light source to the detector between the measurement light and the reference light interfering portion depends on the angle of the measurement light incident on the test sample and the measurement light exiting from the test sample. It corresponds to different scanning coordinates ζ. In other embodiments of the invention, scanning the position of the reference mirror 116 relative to the reference objective 114 (instead of scanning the test sample 112 relative to the measurement objective 110) achieves the same result. Can do.

This difference in how the OPD varies with the scan coordinate ζ introduces a limited coherence length into the interference signal measured at each pixel of the detector. For example, (a function of scan coordinate) interference signal is usually modulated by an envelope having a spatial coherence length of approximately λ / ² (NA) 2. Where λ is the nominal wavelength of the illumination light and NA is the numerical aperture of the measurement and reference objective. As described further below, the modulation of the interference signal provides angle dependent information about the reflectivity of the test sample. In order to increase the limited spatial coherence, the objective lens in the scanning interferometer preferably defines a large numerical aperture, for example a numerical aperture greater than 0.7 (or more preferably greater than 0.9). ).

  The interference signal can be further modulated with a limited temporal coherence length associated with the spatial bandwidth from the illumination source. However, for the purposes of this description, it is assumed that the illumination source is nominally monochromatic and that the temporal coherence limit is small relative to the limited spatial coherence. For example, the illumination source may have a bandwidth that is less than about 2% of its center wavelength.

  Referring again to the Linnik interferometer of FIG. 1, the measurement objective 110 illuminates the surface of the test sample and observes the surface over a range of incident angles φ. Now, assuming a monochromatic illumination, the interference effect will be calculated mathematically using a simplified model. It will then be explained how the optical properties of the sample surface can be found by mathematically resolving the interference fringes into their angle dependent contribution.

  The complex amplitude reflectivity of the surface of the test sample is z (φ), and the corresponding intensity reflectivity Z (φ) is as follows.

  The phase change upon reflection (PCOR) for the sample surface is as follows.

Where “arg” in equation (2) yields the phase of the complex amplitude reflectivity.
In a simplified scalar (non-polarized) model that considers the interference effects individually for each angle of incidence, the interference fringes for a single sample point or camera pixel are proportional to:

Where ζ is the scanning position (moved by PZT) and h is the height profile of the sample surface. The parameters R ₀ (φ), V ₀ (φ), and α ₀ (φ) are DC levels, contrasts, and phase values that are independent of the test sample 112 and that characterize the interferometer optics including the reference mirror 116. is there. As described further below, some calibration procedure identifies these parameters using known artifacts of known optical properties. The R ₀ (φ), V ₀ (φ), and α ₀ (φ) parameters can also include visual field dependence to adjust the optical properties of the instrument, if desired.

  The spatial frequency K (φ) of the interference effect decreases as a function of the angle φ according to the following formula:

  Where λ is the illumination wavelength and the measurement light is assumed to be reflected from the test sample (ie, the measurement light exits the test sample at an angle equal to the angle at which it is incident on the test sample). Equation (4) shows that when scanning is performed, the measurement light (or reflected light) propagates over a range of angles, so that the OPD between the interfering portions of the measurement light and the reference light is the test sample. It is based on the fact that it corresponds to different scanning coordinates ζ depending on the angle of the measuring light incident on it. As a result, equation (4) shows a unique relationship between the spatial frequency of the interference signal and the incident angle.

  Assuming that the light from the light source is spatially perfectly incoherent across the pupil plane and is monochromatic, the net effect of all angle-dependent contributions on the interference phenomenon is the incoherent overlap. Is given by the integral of the sum.

However, φ _MAX = arcsin (NA), and in the following example, the following weight function is used.

  The weight function is suitable for the pupil plane illuminated uniformly with light, which is evident when considering the diagram shown in FIG. 2 (in which the angle is represented by ψ, not φ). )

  For each pixel, the electronic camera and computer control measure the interference fringes I (ζ, h) over a range of scan positions ζ. Height h and effective reflectivity z (φ) vary across the field of view and may vary from pixel to pixel.

  The inherent relationship between the spatial frequency and the incident angle provides a means to recover the individual contribution g (φ, ζ, h) to the integrated fringe I (ζ, h). The first step is to perform complete fringe resolution, for example by Fourier transform.

  Depending on the practical requirements of some limited scan, the integration over all ζ in equation (7) has a limited range of values, including the degree of interference signal needed to obtain an accurate result. Will be terminated. Any other transform that similarly resolves the interference fringes can also be used. The transformation to the spatial frequency domain is generally called frequency domain analysis (FDA).

  The decomposition q [K (φ), h] can be interpreted as follows. Since the zero spatial frequency or DC term cannot be separated as a function of the angle φ, the following equation holds:

  For all other spatial frequency components having a spatial period much smaller than the range that can actually be in the integral, the magnitude and complex phase of q [K (φ), h] are It is as follows.

In one embodiment of the present invention, the optical system characteristics α ₀ (φ), P (φ), V ₀ (φ) are as shown in the text with conventional calibration, eg, Equation (3): Identified by known artifact samples. For example, the measurement can be performed on a test sample having a known surface height and reflectivity so that the properties of the optical system can be derived from equations (9) and (10). If the properties of the optical system are predetermined, Equations (9) and (10) give the surface height and the two optical properties Z (φ) and α _Z (φ) of the surface over the range of incident angles φ Give information about. The optical properties Z (φ) and α _Z (φ) are optical properties themselves that are often tied to specific surface parameters such as thin film thickness by fundamental principles, such as known optical properties of materials and thin films. . Therefore, the three parameters along with the surface height should be adjusted to best fit the measured phase α _Q (φ, h) and magnitude Q (φ, h) of q [K (φ), h]. Can do.

  As an example, consider the thin film structure of FIG. The effective reflectivity of this structure is given by:

Where r ₁ (φ) and r ₂ (φ ′) are the reflectivities of the upper and lower surfaces, respectively, and φ ′ is the angle of incidence on the lower surface calculated from φ and Snell's law. Thin film equation (11) produces a pronounced interference effect that is highly dependent on K (φ).

To quantitatively illustrate this example, a 1.8 micron thin film of silicon dioxide (SiO ₂ ; refractive index n ₁ = 1.46) on silicon (Si; refractive index n ₂ = 3.96 + 0.03i) And an illumination wavelength of 550 nm. The effective reflectivity z (φ) is obtained from the formula (11) and the Fresnel formula for obtaining the interface reflectivity. By scanning the sample surface with respect to the interference objective lens, a signal as in FIG. 4 is generated. For comparison, FIG. 5 shows a simulated interference pattern I (ζ, h) for a simple single-surface SiO ₂ sample (ie, a thick sample of SiO ₂ without a thin film layer).

  After acquiring the data, the computer converts a signal similar to the signal of FIG. 4 into the frequency domain for each image pixel. Due to actual changes in surface topography, optical system parameters, and thin film thickness, the signal and conversion may vary from pixel to pixel. FIG. 6 shows the magnitude (in this case amplitude) of the contribution to each spatial frequency component to the signal of FIG. This result shows a very distinctive feature compared to the frequency domain size shown in FIG. 7 produced by a simple single surface structure with the interference signal shown in FIG.

For example, using FIG. 7 as a calibration and comparing FIG. 6 to FIG. 7, the presence of a thin film is clearly identified. Further, assuming that the properties of SiO ₂ and Si are known, the computer can determine, for example, the thickness of the thin film by comparing FIG. 6 with theoretical expectations based on the effective reflectivity of the sample. it can. This is illustrated by FIG. 8, which compares the expected results of three different thin films, only one of which (1.80 μm) is in good agreement with the Fourier transformed interference data of FIG.

Similar analysis is also useful in the case of the phase of the Fourier transform. 9 and 10 show the difference between a thin film structure and a simple homogenous single surface sample. The obvious non-linearity in FIG. 9 is a clear feature of the thin film effect. Here again, the comparison between measurement and theory gives important thin film information based on equation (10). Further, using the thickness information derived from the amplitude information, α _Z (φ) is identified from Equation (11), and is used in Equation (10) to determine the change in surface height h in various pixels. It can be pulled out.

  In other embodiments, a different interferometry system than in FIG. 1 can be used to provide the scanned interference data I (ζ, h) at each pixel of the camera. For example, a mirrow interferometer as shown in FIG. 11 can be used for the interferometry system.

  Referring to FIG. 11, the light source module 205 provides illumination light 206 to a beam splitter 208 that directs the light to a mirro interference objective lens assembly 210. The assembly 210 includes an objective lens 211, a reference plane 212 having a reflective coating at its small central portion and defining a reference mirror 215, and a beam splitter 213. In operation, the objective lens 211 focuses the illumination light through the reference plane 212 toward the test sample 220. The beam splitter 213 reflects the first portion of the focused light to the reference mirror 215 to define the reference light 222, and transmits the second portion of the focused light to the test sample 220 to define the measurement light 224. Thereafter, the beam splitter 213 recombines the measurement light reflected (or scattered) from the test sample 220 with the reference light reflected from the reference mirror 215, and the light obtained by combining the objective lens 211 and the imaging lens 230. To cause interference on a detector (eg, a multi-pixel camera) 240. Similar to the system of FIG. 1, the measurement signal (s) from the detector is sent to a computer (not shown).

  The scan of the embodiment of FIG. 11 includes a piezoelectric transducer (PZT) 260 coupled to a mirro interference objective lens assembly 210, which completes the assembly 210 relative to the test sample 220 along the optical axis of the objective lens 211. And scanning interferometry data I (ζ, h) is provided at each pixel of the camera. Alternatively, as indicated by the PZT actuator 270, the PZT can be coupled to the test sample, rather than the assembly 210, to provide relative movement therebetween. In yet another embodiment, scanning can be provided by moving one or both of the reference mirror 215 and beam splitter 213 relative to the objective lens 211 along the optical axis of the objective lens 211.

  The light source module 205 is disposed within the spatially extended light source 201, the telescope formed by the lenses 202 and 203, and the front focal plane of the lens 202 (which coincides with the rear focal plane of the lens 203). And a diaphragm 204. This arrangement images a spatially extended light source onto the pupil plane 245 of the Milo interference objective assembly 210, which is an example of imaging with Koehler illumination. The aperture size controls the size of the illumination field on the test sample 220. In other embodiments, the light source module can include an array in which a spatially extended light source is imaged directly onto the test sample, which is known as imaging with critical illumination. Either type of light source module can be used with the Linnik scanning interferometry system of FIG.

In yet another embodiment, the scanning interferometer can be of the Michelson type.
In yet another embodiment of the present invention, a scanning interferometry system is used to identify angle-dependent scattering or diffraction information, ie information for scattered light measurements (scatterometry), for a test sample. Can do. For example, a scanning interferometry system can be used to illuminate a test sample with a test incidence over only a very narrow range of incidence angles (eg, generally normal incidence or otherwise collimated incidence) It can then be scattered or diffracted by the test sample. The light emitted from the sample is imaged on the camera and interferes with the reference light as described above. As with the reflected light of the previously described embodiment, the spatial frequency of each component in the scanning interferometry signal will depend largely on the angle of the test light exiting from the test sample. For near normal incidence, the spatial frequency varies according to the following equation:

The equation differs from equation (4) by a factor of 2 because of normal incidence.
However, there is no change in other parts of the mathematical analysis. Scanning interferometry data I (ζ, h) from a scattered or diffracted test sample is analyzed according to equations (7) to (10), and the test sample Angle-independent phase and amplitude scatter / diffraction coefficients can be provided. Thus, by scanning in the vertical direction (ie, scanning along the optical axis of the objective lens) and subsequent Fourier analysis, it is possible to exit without directly using the rear focal plane of the objective lens or imaging. It becomes possible to measure diffracted and / or scattered light as a function of angle to Furthermore, as described above, the angular dependence of such optical properties can be determined locally across the area of the test sample based on the resolution of the imaging system and the camera pixel size. For example, to provide illumination that is approximately normal, the light source module may image a point light source on the pupil plane or otherwise reduce the degree to which the illumination light occupies the numerical aperture of the measurement objective. Can be configured. Scattered light measurement techniques are used to elucidate discrete structures within the sample surface that can diffract and / or scatter light at larger angles, such as grating lines, edges, or overall surface roughness May be useful to.

  In the above embodiment, the polarization state of the light on the pupil plane is random, i.e. both s (orthogonal to the entrance plane) and p (orthogonal to the entrance plane) polarizations are of approximately equal amounts. It is assumed. It can be realized by a radial polarizer placed in the pupil plane (for example the rear focal plane of the measurement objective in the case of a Linnik interferometer and the rear focal plane of the common objective in the case of a Millo interferometer). Other polarizations are also feasible, including pure s-polarization. Such radial polarization is shown in FIG. Other possible polarizations include radial p-polarization, circular polarization, and modulation (e.g., two states, one state follows the other) for ellipsometric measurements. In other words, the optical properties of the test sample can be elucidated not only with respect to its angular dependence, but also with respect to its polarization dependence or with respect to the selected polarization. Such information can also be used to improve the accuracy of thin film structure characterization.

  In order to implement such ellipsometric measurements, the scanning interferometry system can include a fixed or variable polarizer in the pupil plane. Referring again to FIG. 11, for example, a mirro-type interferometry system uses a polarizing optical system in the pupil plane to select the desired polarization for light incident on and emitted from the test sample. 280. Furthermore, the polarization optics can be made configurable to change the selected polarization. The polarization optics can include one or more elements including a polarizer, a waveplate, an apodization stop, and / or a modulation element to select a given polarization. Furthermore, the polarization optics can be fixed, structured, or reconfigurable to generate data similar to an ellipsometer. For example, a first measurement for s-polarization is performed on a radially polarized pupil plane, and then a measurement for p-polarization is performed on a radially polarized pupil plane. In another example, a pupil plane apodized with linearly polarized light, eg, a slit or optical wedge that can be rotated in the pupil plane to induce any desired linear polarization state to the object, or a liquid crystal display A screen whose configuration can be changed can be used.

Furthermore, polarization optics can provide variable polarization across the pupil plane (eg,
By providing multiple polarizers or spatial modulators). Thus, for example, the polarization state can be “tagged” according to spatial frequency by providing different polarizations for incident angles greater than shallow angles.

  In yet another embodiment, selectable polarization can be combined with phase shift as a function of polarization. For example, a polarizing optical system can include a linear polarizer placed in the pupil plane followed by two wave plates (eg, 1/8 wave plate) in opposite quadrants of the pupil plane. . As a result of the linear polarization, a maximum range of polarization angles is generated with respect to the entrance surface of the objective lens. If the waveplate is aligned such that, for example, the dominant s-polarized light has a certain phase shift, both radial s-polarized and p-polarized light are present simultaneously, but the interferometer Phase shift relative to each other, for example by π, so that effectively detects the difference between these two polarization states as the fundamental signal.

  As described above, by arranging the polarization optical system in the pupil plane, various types of angle-resolved types of polarized light can be measured. However, in yet another embodiment, the polarization optics can be located elsewhere in the apparatus. For example, linear polarization can be achieved anywhere in the system.

  In yet another embodiment, any of the reflected light measurement, scattered light measurement, and ellipsometry techniques described above are repeated in sequence for different wavelengths to provide the wavelength dependence of the sample optical properties of interest. Can do. Such information can be used to match more complex surface models.

  Other embodiments of the invention can include broadband illumination. For example, the illumination can be broadband, as is common in white light interference microscopes, for example. This increases the amount of information that a computer can find to best fit a complex surface model.

  The light source for the scanning interferometry system can be, for example, any of a laser, laser diode, light emitting diode, filtered incandescent light source, and arc lamp.

  The above methods and systems may be particularly useful for semiconductor applications. Yet another embodiment of the invention involves applying any of the above measurement techniques to address any of the semiconductor applications described below.

  In the semiconductor industry, there is currently a great interest in making quantitative measurements of surface topography. Due to the small size of normal chip features, the equipment used to make these measurements typically must have a high spatial resolution both parallel and perpendicular to the chip surface. Engineers and scientists utilize surface topography measurement systems for process control and to detect defects that occur in the manufacturing process, especially as a result of processes such as etching, polishing, cleaning and patterning.

In order to be particularly practical in process management and defect detection, the surface topography measurement system has a lateral resolution that corresponds to the lateral size of normal surface features and a longitudinal direction that corresponds to the lowest allowed surface step height. Must have resolution. This typically requires a lateral resolution of less than 1 micron and a longitudinal resolution of less than 1 nanometer. Also, such a system can damage the surface even if it contacts or does not contact the surface of the chip to avoid changing the surface of the chip or introducing defects. It is preferable to perform the measurement without applying force. In addition, it is well known that the effects of numerous processes used in chip formation are highly dependent on local factors such as pattern density and edge proximity, so surface topography measurement systems have high measurement throughput, and It is also important to have the ability to sample finely over a large area within a region that can contain one or many surface features of interest.

  It has become common for chip manufacturers to use so-called “dual damascene copper” wiring processes to form electrical wiring between various parts of the chip. This is an example of a process in which features can be extracted effectively using a suitable surface topography system. The dual damascene process consists of five parts: (1) Interlayer dielectric (ILD) deposition, where a layer of dielectric material (eg, polymer or glass) is deposited on the surface of a wafer (including multiple individual chips). (2) Dielectric layer is polished to create a smooth surface suitable for high precision optical lithography, chemical mechanical polishing (CMP), (3) narrow extending parallel to the wafer surface A combination of lithographic patterning and reactive ion etching steps in which a complex network is created that includes trenches and small vias extending from the bottom of the trench to the lower (previously defined) conductive layer, (4) As a result, a combination of metal deposition steps, filled with copper until the trenches and vias overflow, (5) excess copper is removed, dielectric Surrounded by a material, a trench filled with copper (and possibly vias) leaving network consisting, it is considered to have a final chemical mechanical polishing (CMP) step.

  Typically, the thickness of copper in the trench area (i.e., the depth of the trench) and the thickness of the dielectric surrounding it is in the range of 0.2 to 0.5 microns. The width of the resulting trench can have a range of 100-100,000 nanometers, and the copper regions in each chip can be a regular pattern, such as an array of parallel lines in some regions. In other regions, there may be no obvious pattern. Similarly, in some areas, the surface can be densely covered with copper areas, while in other areas the copper areas may be sparse. The polishing rate, and hence the thickness of copper (and dielectric) left after polishing, depends on the polishing conditions (eg, pad pressure and polishing slurry composition) and the local fine arrangement of copper and the surrounding dielectric regions (ie , Orientation, proximity and shape) is important to understand in a large and complex manner.

  This “position dependent polishing rate” is known to cause a variety of surface topographies at many lateral length scales. For example, it is generally the case that a chip placed near the edge of a wafer is polished in a shorter time than a chip placed near the center, and is thinner than the desired thickness near the edge and thicker than the desired thickness at the center. It will mean that the region is created. This is an example of “wafer scale” process non-uniformity, that is, non-uniformity that occurs on a length scale corresponding to the diameter of the wafer. It is also known that regions with high density copper trenches are polished faster than near regions with low copper wire density. This results in a phenomenon known as “CMP-induced erosion” in dense copper areas. This is an example of “chip scale” process non-uniformity, that is, non-uniformity that occurs in a length scale that corresponds to (and may be much smaller) the linear dimensions of a single chip. . Another type of chip scale non-uniformity known as “dishing” occurs within a single copper filled trench region (it tends to polish faster than the surrounding dielectric material). . For trenches with a width greater than a few microns, dishing results in severe effects because the affected line will later exhibit excessive electrical resistance and may lead to chip breakage.

Wafer and chip scale process non-uniformities caused by CMP are inherently difficult to predict and they tend to change over time as conditions within the CMP processing system change gradually. In order to ensure that any inhomogeneities are within acceptable limits, process engineers can use many and many different locations to effectively monitor and appropriately adjust process conditions. It is important to frequently perform non-contact surface topography measurement of the chip. This can be achieved using the embodiments of interferometry techniques described above.

  More generally, the interferometry techniques described above can be applied to any of the following surface analysis problems: simple thin films, multilayer thin films, sharp edges that cause diffraction or other complex interference effects, and Depends on surface features, sub-resolution surface roughness, sub-resolution surface features, for example, grooves shorter than wavelength, dissimilar materials, surface polarization on a surface that is otherwise smooth It can be used for warping, vibrations or movements of surfaces or deformable surface features which cause perturbations of properties and consequently interference phenomena depending on the angle of incidence. In the case of a simple thin film, the target variable parameter can be the thin film thickness, the refractive index of the thin film, the refractive index of the substrate, or some combination thereof. For example, in the case of dissimilar materials, the surface can include a combination of thin films and solid metals, and operations that adapt angle-dependent surface properties to a library of theoretical predictions that include both surface structure types. And a thin film or solid metal will be automatically identified by matching against the corresponding interference intensity signal.

  Any of the computer analysis methods described above can be implemented in hardware or software, or a combination of both. The method can be implemented in a computer program using standard programming techniques in accordance with the methods and figures described herein. Program code is applied to the input data to execute the functions described herein and generate output information. The output information is provided to one or more output devices such as a display monitor. Each program can be implemented in a high level procedural or object oriented programming language for interacting with a computer system. However, if desired, the program can be implemented in assembly or machine language. In either case, the language can be a compiled or translated language. In addition, the program can be executed on a dedicated integrated circuit that is programmed to serve that purpose.

  Each such computer program is read by a computer and a storage medium readable by a general purpose or special purpose programmable computer for configuring and operating the computer in performing the procedures described herein. Alternatively, it is preferably stored in a device (eg, ROM or magnetic diskette). Computer programs can also reside in cache or main memory during program execution. The analysis method can also be realized as a computer-readable storage medium configured by a computer program, in which case the computer is specific and predetermined by the storage medium configured as such. Operating in a manner and performing the functions described herein.

  A number of embodiments of the invention have been described. Nevertheless, it will be understood that various modifications can be made without departing from the spirit and scope of the invention.

Schematic of the Linnik scanning interferometry system. The figure which shows illuminating a test sample through an objective lens. FIG. With fringes I (ζ, h) for the structure shown in FIG. 3 laminated with 1.8 μm SiO ₂ on Si, simulated using 550 nm monochromatic light and a 0.9 NA Linnik objective. The figure shows a mixture of interference signals from both surfaces. FIG. 5 is a diagram of simulated interference fringes I (ζ, h) for a simple single surface SiO ₂ sample (ie, no film is present) for comparison with FIG. FIG. 4 shows the magnitude Q (φ, h) of the Fourier transform of the signal of FIG. 4 produced by scanning the thin film structure of FIG. 3 in the vertical direction, the spatial frequency is a figure related to the incident angle according to equation (4). . FIG. 5 shows the Fourier transform magnitude Q (φ, h) of the signal of FIG. 5 for a single surface sample, where the magnitude increases at the lower spatial frequency as a result of increased reflectivity at shallow angles of incidence. Is the graph. P (φ) V ₀ (φ) (Z (φ)) ^1/2 for the thin film structure with SiO ₂ on Si in FIG. 3 for three thin film thicknesses increasing in thickness by 0.02 μm. A graph comparing the expected results of (see equation (9)). The phase α _Q (φ, h) as a function of spatial frequency for the signal of FIG. 4 generated by scanning the thin film structure of FIG. 3 in the vertical direction, and the spatial frequency is the angle of incidence according to equation (4) FIG. 11 is a graph showing significant non-linearities as well as phase gradients compared to the simpler single surface reflection of FIG. FIG. 10 is a graph of phase α _Q (φ, h) as a function of spatial frequency for the signal of FIG. 5 for a single surface pattern for comparison with FIG. 1 is a schematic diagram of a Miro type scanning interferometry system. The figure which shows the radial polarization of a pupil surface.

Claims (46)

By making the test light incident on the test object (112) as incident light, the outgoing light emitted from the test object (112) is referred to on the detector (120) over a certain range of the angle (φ) of the outgoing light . An interferometry method that forms an image so as to interfere with light (106), wherein the interferometry method comprises:
Generating the test light and the reference light (106) from a common light source (102);
An optical path length difference changing step of changing an optical path length difference between the emitted light from the light source (102) to the detector (120) and the reference light (106) in accordance with a change in the angle (φ). When;
The angle (φ) of the optical characteristics of the test object (112) based on the interference between the emitted light and the reference light (106) while changing the optical path length difference at each of the angles (φ). An angle dependency determining step of determining an angle dependency indicating a degree of dependency on
The angle dependence determination step includes
Measuring an interference signal (g) from the detector (120) as the optical path length difference changes simultaneously with the change of the angle (φ);
The interference signal (g), by converting for each said angle (phi) corresponding to the movement to linearly proportional to the optical path length difference scanning coordinates (zeta), previously for the scanning coordinates (zeta) A conversion signal generating step for generating a conversion signal (I) that depends on a conjugate variable (K) that is directly mapped to the incident light or the emitted light. The detector (120) is a camera having a plurality of detector elements,
The imaging includes imaging the emitted light exiting from various locations on the test object (112) to corresponding locations on the camera (120),
The interferometer method according to claim 1. The angular dependence determining step comprises determining the angular dependence for different locations of the test object (112);
The interferometer method according to claim 1. The angular dependence is related to the optical property that varies as a function of the incident angle (φ) of the incident light,
The interferometer method according to claim 1. The interferometry method further comprises:
Illuminating a plurality of locations on the test object (112) with the test light such that the test light is incident on a plurality of locations over a range of the incident angle (φ).
The interferometer method according to claim 4. A common objective lens (211) is used for the irradiation and the imaging.
The interferometer method according to claim 5. The light source (102) is a spatially extended light source,
The interferometer method according to claim 1. The incident angle (φ) range corresponds to a numerical aperture greater than 0.7,
The interferometer method according to claim 5. The incident angle (φ) range corresponds to a numerical aperture greater than 0.9;
The interferometer method according to claim 5. The angular dependence is related to a change in optical properties as a function of the angle of the emitted light scattered from the test object (112).
The interferometer method according to claim 1. The illumination further includes illuminating a plurality of locations on the test object (112) with the incident light having a uniform angle of incidence (φ) on the test object (112),
The imaging includes imaging the emitted light scattered from each location of the test object (112) over a range of angles to a corresponding location on the detector (120).
The interferometer method according to claim 1. A common objective lens (211) is used for the irradiation and the imaging.
The interferometer method according to claim 11. The light source (102) is a point light source.
The interferometer method according to claim 1. The incident angle (φ) range corresponds to a numerical aperture greater than 0.7,
The interferometer method according to claim 11. The incident angle (φ) range corresponds to a numerical aperture greater than 0.9;
The interferometer method according to claim 11. The imaging further includes polarizing the test light at a pupil plane (124) of an optical system related to the imaging.
The interferometer method according to claim 1. The interferometry method further comprises:
Irradiating the test object (112) with the test light;
Polarizing the test light at a pupil plane (124) of an optical system used to illuminate the test object (112),
The interferometer method according to claim 1. The light source (102) is monochromatic.
The interferometer method according to claim 1. The light source (102) has a central wavelength and a spectral bandwidth of less than 2% of the central wavelength;
The interferometry method according to claim 18. The optical path length difference changing step includes moving the test object (112) with respect to an objective lens (110) used to collect the emitted light.
The interferometer method according to claim 1. A reference mirror (116) is used to reflect the reference light (106),
A reference objective lens (114) is used to focus the reference beam (106) on the reference mirror (116),
The optical path length difference changing step includes moving the reference mirror (116) with respect to the reference objective lens (114).
The interferometer method according to claim 1. The step of changing the optical path length includes moving a beam splitter (104) arranged in a mirro interference objective lens.
The interferometer method according to claim 1. The optical path length difference changing step defines a spatial coherence length,
The optical path length difference for at least one of the angles is varied over a range greater than the spatial coherence length;
The interferometer method according to claim 1. The conjugate variable (K) is a spatial frequency.
The interferometer method according to claim 1. The conjugate variable (K) is a spatial frequency K, and the wavelength of the test light is represented by λ.
The direct mapping between the spatial frequency K and the angle φ is given by K∝ (cos φ) / λ,
The interferometer method according to claim 1. The direct mapping between the spatial frequency K and the angle φ is given by K = 4π (cos φ) / λ,
The interferometer method according to claim 25. The transformed signal (I) gives a direct mapping to the angular dependence;
The interferometer method according to claim 1. The transform corresponds to a Fourier transform;
The interferometer method according to claim 27. The optical property is related to the complex reflectivity (z) of the test object (112);
The interferometer method according to claim 1. The optical properties are related to the magnitude of the complex reflectivity (z);
30. The interferometer method according to claim 29. The optical property is related to the phase of the complex reflectivity (z);
30. The interferometer method according to claim 29. The angle dependency is determined in advance by interference between the emitted light and the reference light (106) when the optical path length difference is changed for each angle (φ), and an optical system related to the imaging in advance. Determined based on the angle-dependent characteristics determined,
The interferometer method according to claim 1. The interferometry method further comprises:
Determining a surface height profile of the test object (112) based on interference between the emitted light and the reference light (106) when the optical path length difference is changed,
The interferometer method according to claim 1. The interferometry method further comprises:
Comparing the change in angular dependence determined from the interference between the outgoing light and the reference light (106) with the change in angular dependence of the outgoing light of the model for the test object (112). including,
The interferometer method according to claim 1. The test object (112) includes at least one thin film located on a substrate;
35. The interferometer method of claim 34. The interferometry method further comprises:
Based on the comparison, including a thickness determination step of determining the thickness of the thin film,
36. The interferometry method according to claim 35. The optical properties include the magnitude of the angular dependence of the complex reflectance (z) of the test object (112);
The thickness determination step is based on comparing the magnitude of the angular dependence of the complex reflectance (z) with the magnitude of the angular dependence of the emitted light.
The interferometry method according to claim 36. The interferometry method further comprises:
Determining a surface height profile of the test object (112) based on the comparison;
The interferometry method according to claim 37. The optical property further includes an angle dependent phase of the complex reflectivity (z),
The determination of the surface height profile compares the determined thickness of the thin film; the angle dependent phase of the complex reflectivity (z); and the angle dependent phase of the surface high profile at the determined thickness. Based on
39. An interferometry method according to claim 38. The test light and the reference light (106) have a first wavelength,
The interferometry method further repeats imaging, changing the optical path length difference, and determining the angular dependence in the case of test light and reference light (106) having a second wavelength different from the first wavelength. Including,
The interferometer method according to claim 1. A common light source (102) for generating test light and reference light (106);
A detector (120);
A scanning stage (126) for mounting a test object (112);
A scanning interferometer comprising an electronic processor (128) coupled to both the detector (120) and the scanning stage (126) ,
The scanning interferometer causes the outgoing light emitted from the test object (112) to be incident on the test object (112) as incident light, by using a certain range of the angle (φ) of the outgoing light. Is configured to image to interfere with the reference beam (106) on the detector (120)
The scanning stage (126) corresponds to an optical path length difference between the emitted light from the light source (102) to the detector (120) and the reference light (106) according to a change in the angle (φ). Is configured to define a scanning coordinate (ζ) that moves linearly proportional to the optical path length difference,
The electronic processor (128) changes the optical path length difference at each of the angles (φ), and based on the interference between the emitted light and the reference light (106), the optical of the test object (112) properties, is configured to determine the angular dependence showing the dependence degree of the to angle (phi),
The detector (120) is configured to measure an interference signal (g) in response to a change in the optical path length simultaneously with a change in the angle (φ),
Said electronic processor (128), said interference signal (g), by converting for each said angle (phi) the scan coordinates corresponding to (zeta), the incident light and the respect to the scanning coordinates (zeta) A scanning interferometer, characterized in that it is configured to generate a transformed signal (I) that depends on a conjugate variable (K) that is directly mapped to outgoing light. The scanning interferometer further includes:
An objective lens (110) arranged to collect the emitted light;
At least one polarization optical system (280) disposed on the pupil plane (124) of the objective lens (110).
The scanning interferometer of claim 41. The polarizing optical system (280) provides polarized light that varies across the pupil plane (124);
43. A scanning interferometer according to claim 42. The polarizing optical system (280) includes a polarizer and at least one wave plate.
43. A scanning interferometer according to claim 42. The polarizing optical system (280) includes two wave plates arranged at different positions on the pupil plane (124).
45. A scanning interferometer according to claim 44. The light source (102) is a monochromatic light source,
The scanning interferometer of claim 41.

Images