Vertical resolution metrology with asymmetry sensitivity

Abstract

A system and method for OCD metrology is provided that includes setting an interferometer mirror of the system at each of a plurality of positions z, and measuring a spectrum of the interferometer signal reflected from a lower surface and an upper surface while applying a first setting of an asymmetry sensitive measurement parameter, thereby creating a set of a plurality of spectra. Subsequently, a function representing a coherent portion of the reflected signal is determined and transformed to the time domain to generate a first signal feature representing the coherent portion of the reflected signal. Subsequently, the process is repeated for a second setting of the asymmetry sensitive measurement parameter to generate a second signal feature. A measure of the difference between the first signal feature and the second signal feature is indicative of a measure of the asymmetry of the structure.

Description

Technical Field

[0001] This invention generally relates to the field of optical inspection of integrated circuit wafer patterns, and more specifically, to algorithms for measuring wafer parameters. Background Technology

[0002] Integrated circuits (ICs) are produced on semiconductor wafers through multiple steps of depositing, altering, and removing thin layers that accumulate into a stacked structure. These stacked structures (also known as "stacks" or "features") have optical properties similar to diffraction gratings.

[0003] Optical critical dimension (OCD) metrology employs scattering measurements to measure the light radiation reflected from structures formed on a sample (i.e., a silicon wafer) during IC manufacturing. Common scattering measurement methods include spectral reflectance (SR), spectral ellipsometrics (SE), and spectral interferometry (SI). Scattering measurements are generally used in OCD metrology during IC manufacturing to determine whether the wafer pattern is being manufactured with the correct parameters. Measurements can determine the degree of variation in design specifications. Manufacturing agreements can specify permissible deviations from the average value.

[0004] Scattering measurement methods face significant challenges when measuring thick structures, especially those with several reflective surfaces that are clearly separated from each other. A common application with these characteristics is the fabrication of "through silicon vias" (TSVs), which is particularly common for enabling technologies such as chip fabrication.

[0005] TSV structures are etched circular holes that are then filled with metal and used as interconnects. TSVs typically have depths ranging from tens to hundreds of micrometers. Because they are greater than 10 μm, TSVs are considered relatively tall structures in semiconductor manufacturing. Other examples of tall (or “thick”) structures include ultra-thick dielectrics, 3D-NAND memory devices, CMOS image sensors, and a variety of other structures used in advanced semiconductor packaging.

[0006] A range of techniques for addressing the difficulties in characterizing these structures have been described in several international patent applications filed under the name of Nova Measuring Instruments Ltd., specifically the following international patent applications: WO2022162617, “Time-domain optical metrology and inspection of semiconductor devices”; WO2024003758, “Optical critical dimensions (OCD) metrology for thick stacks”; and WO2024100674, “Coherent spectroscopy for…” "TSV (Coherent Spectrometry for TSV)," and all of these international patent applications are incorporated herein by reference. The methods described in these publications analyze signals reflected from vertically distinct regions of a structure by including a process of filtering the incoherent portions of the signal and then transforming the filtered signal from the frequency domain to the time domain, in order to identify coherent signals with less spectral tails from specific layers of the structure. In the following, such approaches and methods are referred to as "vertical-resolved metrology" (VRM). The previously described VRM techniques can be used, for example, to characterize the total depth of a TSV.

[0007] Another metrological challenge for tall structures involves characterizing the lateral dimensions of the TSV, and in particular, the characterization of structural asymmetry—that is, deviations from the nominal symmetrical design specifications of such structures. Examples of asymmetry include structural tilt in overlay metrology, where asymmetry is introduced by relative offsets between different layers. Additional asymmetry is dimensional asymmetry, where, for example, the cross-section of a cylindrical TSV becomes elliptical.

[0008] Several techniques exist for characterizing asymmetries in thin wafer structures that are typically a few nanometers in height. For example, overlay metrology using image-based overlay (IBM) or diffraction-based overlay (DBO) can identify some asymmetries. For instance, ellipsometric metrology utilizing polarization variations caused by asymmetries is described in, for example, U.S. Patent 10,876,959 filed in the name of Nova Measuring Instruments Ltd.

[0009] However, while these methods can identify and quantify overall symmetry breaking in samples, they cannot identify the precise location or layer of asymmetry, and specifically, they are not designed to identify asymmetries present in high (or “thick”) structures such as TSVs as described above. Even with advancements in VRM, high structural asymmetries (such as those mentioned above) pose a significant metrological challenge because sensitivity to small structural differences is often overwhelmed by other sensitivities. Summary of the Invention

[0010] Embodiments of this invention provide interferometry based on optical path difference (OPD) matching, extending the vertical-resolved metrology (VRM) method for optical critical dimension (OCD) metrology by applying asymmetry-sensitive measurement techniques. The processes described herein allow for the resolution of symmetry deviations with improved signal-to-noise ratio (SNR), while also identifying specific parameters of symmetry breaking. This invention permits the characterization of systematically high structural asymmetries. This type of asymmetry can be particularly pronounced at the bottom of deep structures, where the TSV may, for example, have different widths in different directions. Deterioration of fabrication components within the wafer processing chamber can cause such asymmetries, and timely identification of these problems prevents manufacturing waste and inefficiency. Attached Figure Description

[0011] To better understand the various embodiments of the invention and to illustrate how the invention can be practiced, reference is made to the accompanying drawings by way of example. Structural details of the invention are shown to provide a basic understanding of the invention, and the description in conjunction with the drawings makes it apparent to those skilled in the art how several forms of the invention can be practiced in practice. In the drawings: Figure 1 This is a schematic diagram of a system for optical critical dimension (OCD) measurement according to an embodiment of the present invention, the system including adjustable elements for asymmetry-sensitive measurements; Figure 2A and Figure 2B The temporal-domain plots of a TSV structure with several layers and patterned near the TSV, and coherent reflections from the TSV surface generated by matching the optical path difference (OPD) between the top and bottom surfaces are shown respectively. Figure 3A and Figure 3B The tilt and elliptical asymmetry of the TSV are shown respectively, which may be caused by manufacturing deviations; Figure 4A and Figure 4B This is a schematic diagram of a system for OCD metrology, which employs an adjustable polarization filter for asymmetry-sensitive measurements. Figure 5A and Figure 5BThis is a schematic diagram of a system used for OCD metrology, in which the interferometer mirror support system has an adjustable tilt for performing asymmetry-sensitive measurements; Figure 6A and Figure 6B This is a schematic diagram of a system used for OCD metrology, which employs an adjustable spectrometer aperture setting for asymmetric-sensitive measurements. Figure 7A and Figure 7B This is a schematic diagram of a system used for OCD metrology, which employs adjustable spot offset for asymmetric-sensitive measurement. Figure 8 It is a graph of the time-domain impulse response transformation of the spectral results, indicating the differences between the results obtained using different settings of the asymmetry-sensitive measurement parameters; Figure 9 and Figure 10 This is a schematic diagram of the wafer, showing the ratio of TSV elliptic asymmetry at various locations across the entire wafer; and Figure 11 This is a flowchart depicting a process for characterizing OCD metrology according to an embodiment of the present invention, the process including OPD matching interferometry and one or more asymmetry-sensitive measurements. Detailed Implementation

[0012] Embodiments of the present invention provide systems and methods for optical critical dimension (OCD) metrology based on optical path difference (OPD) matched interferometry, utilizing one or more asymmetry-sensitive measurements.

[0013] Figure 1 This is a schematic diagram of a system 20 for spectroscopic interferometry metrology of wafer 22 according to an embodiment of the present invention, which is used to characterize structural asymmetries, particularly high-structure ones. System 20 can operate within a production line (not shown) for producing and monitoring the structure of wafers. Additional details and basic principles of spectroscopic interferometry (SI) can be found in U.S. Patent No. 10,161,885 and PCT Patent No. PCT / IB2022 / 050774, both of which are incorporated herein by reference under the name of Nova Ltd.

[0014] Wafer 22 typically has multiple surfaces, which are also referred to herein as “layers” or “interfaces.” In the example shown, there are separation distances. H The system is shown as having a higher surface 24 and a lower surface 26. The system is configured to characterize structures extending from the higher surface to the lower surface 26, such as "through silicon vias" (TSVs) 28.

[0015] Light is projected from light source 30, typically having a predetermined frequency range (such as broadband illumination covering the visible light range) or extending into the infrared and / or ultraviolet range. The light is directed towards the structure being measured by a first beam splitter (BS) 32. The light is then split by a second beam splitter 34, with one portion directed towards the sample and the other portion directed towards the interferometer mirror 36.

[0016] The reflections from these two parts of the light (that is, from the sample and from the interferometer mirror) are recombined at the second beam splitter 34 and directed toward the spectrometer 40. The data from the spectrometer can then be processed by the processor 42 to provide the computations described herein for identifying and characterizing structural asymmetries.

[0017] The interferometer mirror can be linearly moved along linear axis 46 to a position "matched" with the reflecting surface of the structure being measured, such that the optical path difference (OPD) between the light traveling to the target surface (i.e., the layer) and the light traveling to the mirror is zero. For example, OPD matching of the mirror with the higher surface involves setting distance B (distance from the mirror to beam splitter 34) equal to distance A (distance from the beam splitter to the higher surface). OPD matching of the lower layer involves setting B to A+. H .

[0018] The optical path length of each path traversed by light is a function of the geometric length and the refractive index of the material in which the light is propagating; that is, the length is measured in relation to light propagation. Within the structure, each electromagnetic eigenmode has a different phase rate, and therefore a different effective refractive index and a different OPD.

[0019] System 20 can be applied to implement vertically resolved metrology (VRM) methods, and further enhanced by applying asymmetric sensitivity measurement techniques. For example, in some embodiments of the invention, as further described below, mirror 36 may have a tilt axis 48 as an additional axis of movement, about which the mirror can rotate for asymmetric sensitivity measurements.

[0020] In additional or alternative embodiments of the invention, as further described below, a polarizing filter (designated polarizer 50) may be located along the path of light, typically between the first and second beamsplitters. The polarization direction may also be adjusted for asymmetry-sensitive measurements. Objective lens 52 can be used to focus light passing through the polarizer onto the sample.

[0021] After reflection, the light from the sample and the mirror is combined by the second beam splitter as described above, and can be collimated by the objective lens before passing through the polarizer and the first beam splitter again to reach the spectrometer. Typically, additional lenses (e.g., tube lenses, not shown) are provided to guide the light. Typically, the aperture 60 also restricts the light entering the spectrometer.

[0022] In another embodiment of the invention, as further described below, the opening 60 can be adjusted to perform asymmetry-sensitive measurements.

[0023] In additional or alternative embodiments, as further described below, alternative lenses of the objective or system (such as lenses of the light source (not shown)) can be adjusted to cause a shift in the light spot at the target surface of the wafer, thereby enabling asymmetry-sensitive measurements. Alternatively, the stage on which the wafer is placed can be moved to cause a shift in the light spot.

[0024] Additional components of system 20 may include an imaging lens, additional polarizing filters, a variable aperture stop, and motors (not shown). The operation of many system components (such as mirrors, beam splitters, filters, aperture stops, and motors) is typically automated by a computer controller (such as processor 42), which may include I / O devices and may also be configured to perform data processing tasks, such as generating scattering measurement data from received spectrometer signals. In typical OCD metrology, the range of light measured can cover the visible spectrum and may also include wavelengths in the ultraviolet and infrared regions. A typical spectral output for OCD metrology may have, for example, 245 data points covering a wavelength range of 200 nm to 970 nm.

[0025] Figure 2A The TSV structure 28 of wafer 22 as described above is shown, with several layers (or "surfaces") of structure 100 adjacent to the top of the TSV structure. Spots of light from a light source are simultaneously reflected from multiple layers of this structure. Figure 2B This illustrates the VRM technology described above (in the background section), where the data determined by OPD matching originates from... Figure 2A The time-domain plot of the coherent reflection of the structure shown. That is, Figure 2B It is a function representing the coherent portion of the reflectance spectrum of a structure (such as a TSV) in the time domain. This function reveals distinct peaks, where the spacing between the top and bottom surfaces varies with the stacking height. H Proportional.

[0026] It should be noted that, given that the x-axis represents the distance traveled by light, the x-axis can be in units of time or distance (such as optical path length, OPL). As shown, multiple peaks appear around the shorter optical path length of the top structure 100. An isolated peak was obtained on the bottom surface of the TSV.

[0027] Figure 3A and Figure 3B An example of a high structure with structural asymmetry defects is shown, specifically, the asymmetry of etched circular hole structures (such as that that may occur in TSV or 3D-NAND channel holes). Figure 3A The structural tilt is shown (which could be a consistent tilt over the through-hole area as shown, or an asymmetry caused near the bottom of the structure). Figure 3B The diagram shows a break in the circular symmetry at the bottom of the structure, meaning that the bottom has different via widths in different directions. Another asymmetry, not shown, is overlay error; the TSV is not centered on its target underlying structure. Asymmetries such as these can be caused by manufacturing deviations, such as uneven distribution of etching plasma within the chamber.

[0028] This invention provides systems and methods for determining whether a high-structure exhibits the aforementioned types of asymmetry. The basic process involves performing separate interferometric measurements while applying two or more different asymmetry-sensitive parameter settings. For each type of asymmetry-sensitive parameter (also referred to herein as a “channel”), the difference between the results when applying different settings produces different interferometric results (also referred to herein as a “signal characteristic”) is an indicator of asymmetry.

[0029] The differences between results from different asymmetry-sensitive settings can also be compared with previously calibrated metrics (such as thresholds) or “feature models” to determine the type and extent of asymmetry. In some cases, measurements from two different settings are approximately equal for a given parameter (i.e., “channel”) when the sample structure is symmetrical (e.g., the TSV has no tilt and / or the TSV bottom is rounded, etc.), and measurements differ when the sample symmetry is broken (e.g., the TSV bottom is elliptical).

[0030] As described above, asymmetry-sensitive parameters may include: 1) polarization of the light source and / or reflected light; 2) modification of the light incident angle, for example, by changing the tilt of the interferometer mirror; 3) modification of the angular span of the light reaching the spectrometer aperture, for example, by blocking a portion of the aperture; and 4) changing the deflection of the light incident on the sample. Methods for adjusting these four types of asymmetry-sensitive parameters will now be described in more detail with respect to the accompanying drawings.

[0031] Figure 4A and Figure 4BThis is a schematic diagram of a system 20 with an adjustable polarizer 50 for performing asymmetry-sensitive measurements. When the incident light is perpendicular to the sample surface, the reflected signal is approximately the same for all incident light polarization planes when examining a symmetrical structure. However, when one side of the structure is asymmetrical, such as when the TSV has an elliptical or sloping bottom critical dimension (BCD), the intensity (i.e., amplitude) of the polarized reflected light changes with the polarization direction. As an example, light polarized in the plane of the major axis at the bottom of an elliptical TSV is reflected with a different intensity than incident light polarized along the minor axis at the bottom of an elliptical TSV. The difference in signal characteristics between reflected signals from different polarizations, isolated by the VRM technique described above, indicates the type and extent of symmetry breaking. Typically, two polarization settings can be used with orthogonal polarization, such as... Figure 4A and Figure 4B The different polarization patterns of the polarization filter 50 in the image are indicated. When the possible asymmetry directions are unknown, additional measurements under different polarization orientations can be performed to characterize the asymmetry. As described below, elliptical asymmetry in BCDs often manifests as the elliptical axis aligning with the radial and tangential directions of the wafer. In this case, polarization in these orthogonal directions is an efficient method for identifying this type of asymmetry.

[0032] Figure 5A and Figure 5B This is a schematic diagram of the incident angle (AOI) asymmetry-sensitive parameter, achieved by adjusting the tilt of the interferometer mirror 36 around the tilt axis 48. Although SI measurements are typically performed perpendicular to the sample incidence, a small tilt can be introduced between the optical axis and the wafer surface. When the target structure is symmetrical, two VRM measurements of a symmetrical structure performed at two different AOIs usually provide the same results. When the sample symmetry is broken (i.e., the target structure is asymmetrical), these measurement results become unequal.

[0033] This type of measurement scheme can be achieved using a tilted mirror, whereby the angular span of the light rays reflected from the mirror differs from that reflected from the sample (the offset is twice the mirror angle). During interference, the two beam portions (from the sample and from the mirror) only... part Interference occurs within the angular span. Therefore, the coherent portion of the collected signal begins at the tilted angular span.

[0034] Alternatively, the incident angle span can be fixed at a non-vertical span, and the sample can be rotated to allow for different measurements at different azimuth angles.

[0035] The tilt angle is typically about 0.5 degrees or less, and in any case, it does not exceed the angle at which light from the mirror no longer reaches the spectrometer aperture. The results can be enhanced by applying polarization (e.g., applying polarization oriented at 45 degrees to the AOI).

[0036] Figure 6A and Figure 6B This is a schematic diagram of a system for OCD metrology, in which the setting of the spectrometer aperture 60 is adjusted for asymmetry-sensitive measurements. The asymmetric pupil surface introduces asymmetry in the angular span of the measurement, which can be achieved, for example, by partially obscuring the numerical aperture of the system. Such limitation can be achieved using a movable aperture and / or shutter under processor control. This asymmetry can be achieved at the illumination source or at the collection point (i.e., the spectrogram aperture as shown), or both.

[0037] Figure 7A and Figure 7B This is a schematic diagram of a system used for OCD metrology, which employs asymmetric-sensitive measurements by adjusting the spot offset (under processor control). Offsetting the incident light spot with a small lateral offset relative to the TSV (with or without additional angular offset) alters the measurement results and the characteristics of the resulting signal. This offset can be achieved by moving the system's lenses or by moving the wafer (as long as the movement maintains at least partial light coverage of the surface of the structure being inspected (e.g., the bottom of the TSV).

[0038] Figure 8 This is a time-domain impulse response transformation graph of the spectral results, indicating the differences between results obtained using different asymmetry-sensitive parameter settings. As described above, using the principle of vertical-resolved metrology (VRM) (explained in the above-cited publications with reference to the flowchart described below), the measured interferometric signal is converted to a time-dependent impulse response domain to isolate coherent signals with different height correlations in the stacked layers of the high-structure.

[0039] Figure 8 The results shown are for structures with a single top surface and a single bottom surface, each indicated by a separate peak. The peaks on the top surface are unaffected by the setting of the asymmetry-sensitive parameter. However, the asymmetry of the bottom surface causes the peaks of the signal acquired from the bottom surface to change according to the setting of the asymmetry-sensitive parameter.

[0040] When comparing two measurements under different asymmetry-sensitive settings, asymmetry-based differences are observed in the reflection characteristics associated with bottom asymmetry. This method can also be used to identify the vertical location of symmetry-broken elements. This can be achieved by transforming the time at which the spectra of the two measurements differ and converting the time difference into a distance metric to identify the location within the structure stack. Therefore, the difference between signal characteristics provides an indication of asymmetry. In some cases, the target for the difference is zero; that is, any deviation may require intervention in the manufacturing process.

[0041] As mentioned above, differences between signal features can also be analyzed using techniques such as calibration, model-based interpretation, machine learning analysis, or any similar interpretation protocol to determine the Target of Interest (POI), specifically, the type and magnitude of structural asymmetry, i.e., its size. Such a size, also referred to herein as a “measure of asymmetry,” can include, for example, the ellipse size of the TSV base. Methods for modeling and calibrating the correlation between measures of difference and measures of asymmetry can rely on initial analysis or “training” of a model of the asymmetry effect. Differences between features generated by different asymmetry sensitivity settings can also be quantified as relative or absolute differences in peak intensity, amplitude, or pattern. The level of asymmetry can then be compared to a previously determined threshold or scale to indicate the size of the asymmetry or other measures of asymmetry, such as its severity. Asymmetry greater than a preset level (or, in some cases, zero) may require, for example, human operator intervention and manufacturing process adjustments.

[0042] To calibrate or model the measure of difference against a real measure of asymmetry, the measure of asymmetry can first be determined using other techniques. One technique that can be used is to apply a high-energy electron beam tool to measure the bottom of the tall structure. A possible additional technique is critical-size small-angle X-ray scattering (CD SAXS). To measure the ellipticity and tilt of a tall structure (such as a TSV), after obtaining data on the measure of difference to be correlated, the TSV can be filled with copper (Cu), for example, and subsequently, after bonding the top side of the wafer, the back side of the wafer can be polished to expose the bottom of the TSV. The exposed filling can then be measured, for example, using critical-size scanning electron microscopy (CD SEM). After correlating the measure of difference with the measure of asymmetry, the more inexpensive and efficient interferometric measurement methods described herein can then be used in production.

[0043] Figure 9 and Figure 10 This is a schematic diagram of wafer 22, illustrating the TSV BCD elliptic asymmetry at various locations across the wafer. Significant sensitivity benefits can be achieved by appropriately setting the asymmetry sensitivity parameter when the parameter of interest (POI) (such as the TSV bottom shape) has known, expected behavior. Figure 9 The theoretical deviation of the bottom shape of the high-structure relative to the circular shape is shown due to etching issues. As indicated, the expected asymmetry is along the radial axis of the wafer.

[0044] To improve asymmetry sensitivity, asymmetry-sensitive parameters (such as polarization or AOI settings) can be configured to match the parallel and perpendicular axes of the expected elliptical asymmetry. Similarly, measurements can be optimized for appropriate mirror tilt, angular span offset, or measurement position offset based on the expected structural tilt direction. The difference in reflection intensity between radial and tangential asymmetry-sensitive settings indicates the presence of anticipated, etching-related deviations.

[0045] Figure 10 Test results for TSV asymmetry on a wafer fabricated with multiple TSVs are shown, where the polarizer is oriented in both the vertical and horizontal directions in each set of measurements performed at each measurement location. Shading indicates the relative intensity of the bottom peak reflection, i.e., the intensity difference between the two polarization directions. As shown, the asymmetry is readily apparent at the ends of the horizontal and vertical axes of the wafer. At a 45-degree offset, the asymmetry is not significant given that the ellipse radii are equal at ±45 degrees. As discussed above, given the tendency for asymmetry along these axes at any given location on the wafer to be attributable to etching deviations, asymmetry-sensitive settings along the radial and tangential directions of the wafer would be a better indicator of asymmetry.

[0046] Figure 11 It is a flowchart depicting the process for characterizing asymmetry in OCD metrology, which includes OPD-matched interferometry using VRM filtering and one or more asymmetry-sensitive measurements.

[0047] The process includes performing a set of OCD measurements at step 510 using a first setting of asymmetry-sensitive parameters. This involves measuring multiple interferometer spectra by simultaneously applying the first setting of the asymmetry-sensitive parameters while positioning the interferometer mirrors at different locations z around the OPD of the surface under examination. Alternatively, a combination of asymmetry-sensitive parameters can be set as a first combination of values ​​(and subsequently as a second combination at a later step 530 described below). For typical asymmetry studies, the relevant surface will be the bottom of a tall structure (such as a TSV). Asymmetry-sensitive parameters can be, for example, polarization direction, mirror tilt, aperture offset, or spot offset, or any combination of these parameters.

[0048] All OPD measurements should be taken across the entire coherence range of the target surface, i.e., within the "coherence length" of the light source (so that a "partial coherence" factor is not required). For multiple distinguishable height differences between surfaces, each distinct surface can be measured using individual OPD matching.

[0049] In step 520, the measured multiple spectra are filtered using the VRM technique described above, thereby fitting the signal to... Imeasured A mathematical equation, which can be defined, for example, as:

[0050] in, k It is the wavenumber during each interferometer measurement (= ); z i yes i The position of the reflector in each measurement; ( k ) are the corresponding reflection pairs. tb (Top - Bottom) tm (Top - Reflector) and bm Wavelength-dependent reflection phase difference (bottom-reflector); H It refers to the geometric differences between interfaces; n is the refractive index of the structure; r m , r t , r b These are the reflectivities of the top and bottom surfaces of the mirror, respectively. k The function; γ It is the coherence of OPD 2nH; γ 'It is OPD 2 ( nH - z The coherence of ) R It is a real function with an imaginary part; and coherence γ It is defined as the ratio between the measured interferometric term and the expected interference of a fully coherent field (which can be measured, for example, by moving one of the mirrors in a Michelson interferometer and measuring the amplitude of the resulting oscillation).

[0051] Then, the measured spectra are fitted to the desired functional form. I measured ( k, z Then, non-reflective elements can be canceled for all mirror positions. z Related items, leaving out the differences between the top and bottom respectively. z Related functions. z The correlation function is transformed to the pulse time domain, which then allows the identification of coherent reflection signals from the inspected surface for a given asymmetry-sensitive parameter setting, i.e., signal characteristics.

[0052] At step 530, the above steps are then repeated for a second setting of one (or more) asymmetry-sensitive parameters to generate a second signal feature. Additional features may also be generated by repeating steps 510 to 520 for settings of asymmetry-sensitive parameters to provide additional specificity and thereby overcome potential problems such as asymmetry orientation in unknown directions.

[0053] At step 540, a difference or set of differences is determined in the distinguishing features (i.e., the time-domain spectrum). The difference can be between the maximum values ​​of the peaks, or between the maximum values ​​at a given time (i.e., the values ​​of the x-axis parameter). Another measure of difference can be the area under the feature curve, which can be calculated over a given time range. Other relative relationships between features can also be used, such as the ratio between maximum values. The difference between features can also be represented as a difference function, which can be compared to a previously generated difference model.

[0054] At step 550, the calculated difference can be compared with a previously determined threshold or model of asymmetry to generate indicators of asymmetry. These indicators can then be output to the operator, for example, as an alert about a manufacturing problem.

[0055] The advantages of this invention include: High-precision characterization of symmetry-broken points of interest (POIs) buried in the measured structure; Robustness to details and changes in other parts of the stack; Identify the vertical locations where the symmetry of the stack is broken.

[0056] It should be understood that the processing elements shown or described herein are preferably implemented by one or more computers in computer hardware and / or computer software contained in a non-transient computer-readable medium according to conventional techniques, such as employing computer processors, memory, I / O devices, and network interfaces coupled via computer buses or alternative connection means.

[0057] Unless otherwise stated, the terms "processor" or "device" are intended to include any processing device, such as, for example, a processing device including a CPU (Central Processing Unit) and / or other processing circuitry (e.g., a GPU), and may refer to more than one processing device. Various elements associated with a processing device may be shared by other processing devices.

[0058] As used herein, the term "memory" is intended to include memory associated with a processor or CPU, such as, for example, RAM, ROM, fixed memory devices (e.g., hard drives), removable memory devices (e.g., floppy disks, magnetic tapes), flash memory, etc. Such memory can be considered as computer-readable storage media.

[0059] Additionally, the phrase "input / output device" or "I / O device" may include one or more input devices (e.g., keyboard, mouse, scanner, HUD, etc.) for inputting data to the processing unit, and / or one or more output devices (e.g., speaker, display, printer, HUD, AR, VR, etc.) for presenting results associated with the processing unit.

[0060] Embodiments of the present invention may include systems, methods, and / or computer program products. The computer program product may include a computer-readable storage medium (or multiple computer-readable storage media) having computer-readable program instructions thereon for causing a processor to execute aspects of the present invention.

[0061] Computer-readable storage media can be tangible devices that can retain and store instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer-readable storage media includes the following: portable computer floppy disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable optical disc read-only memory (CD-ROM), digital versatile disc (DVD), Blu-ray, magnetic tape, holographic memory, memory sticks, floppy disks, mechanical encoding devices (such as punched cards or raised structures in grooves with instructions recorded thereon), and any suitable combination of the foregoing. As used herein, computer-readable storage media should not be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0062] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a suitable computing / processing device, or downloaded via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network) to an external computer or external storage device. A network adapter card or network interface in each computing / processing device can receive computer-readable program instructions from the network and forward them for storage in a computer-readable storage medium within the suitable computing / processing device.

[0063] Computer-readable program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, Smalltalk, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be to an external computer (e.g., via the Internet through an Internet service provider). In some embodiments, electronic circuit systems including, for example, programmable logic circuit systems, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs) may execute computer-readable program instructions by utilizing state information from the computer-readable program instructions to personalize the electronic circuit system, thereby performing aspects of the invention.

[0064] In the context of describing aspects of the invention herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention, it will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0065] These computer-readable program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions may also be stored in a non-transitory, computer-readable storage medium that can instruct a computer, programmable data processing apparatus, and / or other apparatus to operate in a particular manner, such that the computer-readable storage medium storing the instructions may include an article of manufacture comprising instructions that implement aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0066] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus, or other device, thereby producing a computer-implemented process, such that the instructions, which execute on the computer, other programmable apparatus, or other device, perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0067] Any flowcharts and block diagrams included herein illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of instructions, which may include one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may appear in the order shown herein. For example, two blocks shown consecutively may actually be executed substantially simultaneously, or these blocks may sometimes be executed in reverse order, depending on the functions involved. It will also be noted that each block in the block diagrams and / or flowchart illustrations, and combinations of blocks in the block diagrams and / or flowchart illustrations, may be implemented by a dedicated hardware-based system that performs the specified function or action or executes a combination of dedicated hardware and computer instructions.

[0068] Various embodiments of the invention have been described for illustrative purposes, but are not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles of the embodiments, their practical application, or technical improvements to technologies found in the market, or to enable those skilled in the art to understand the embodiments disclosed herein.

[0069] Example

[0070] The present invention may include the following configuration.

[0071] Example 1 is a method for optical critical dimension (OCD) metrology, used to characterize a structure having at least one lower surface separated from a higher surface. Typically, this structure is a tall structure, such as a TSV. The method includes: a) For each of the plurality of linear positions z of the interferometer mirror of the interferometric measurement system, while applying a first setting of asymmetry-sensitive measurement parameters, the spectra of the interferometer signals reflected from the lower surface and the higher surface are measured to create a set of plurality of spectra, wherein each linear position z is within the coherent optical path difference (OPD) matching range of the lower surface. b) From the set of measured spectra, determine a function representing the coherent portion of the reflected signal, and apply a time-domain transformation to the function to generate a first signal feature representing the coherent portion in the time domain; c) Repeat steps a and b for the second setting of the asymmetric sensitive measurement parameter to generate a second signal feature; d) Calculate a measure of the difference between the first signal feature and the second signal feature, and output the measure of the difference as an indicator of the asymmetry of the structure.

[0072] Example 2 includes the above features, and the light emitted at the light source of the interferometric measurement system has a coherence length smaller than the distance that separates the lower surface from the higher surface.

[0073] Example 3 includes features from Example 1 or 2, where the measure of difference is a function of the difference between the first and second features within the optical path difference (OPD) matching range of the features.

[0074] Example 4 includes features from any of the examples above, and further includes determining the severity of asymmetry by comparing a measure of the difference with a previously calibrated threshold.

[0075] In Example 5, the difference between the first signal feature and the second signal feature can be measured as the difference between the areas under the curves of the corresponding signal features.

[0076] In Example 6, the difference between the first signal feature and the second signal feature can be measured as the difference between the signal amplitudes of the corresponding signal features.

[0077] Example 7 includes features of any of the examples above, and determines one or more size asymmetries by comparing a measure of the difference with a previously determined measure of the difference associated with one or more size asymmetries of the structure via a function or machine learning model.

[0078] Example 8 includes features of any of the examples above, and the asymmetry-sensitive measurement parameter is one or more of the following: the polarization state of the reflected interferometer signal; the angle of incidence (AOI) achieved by the tilting orientation of the mirror; aperture adjustment for partially blocking light emitted at the light source or received at the spectrometer of the interferometric system; and the offset of the light spot focused on the structure by the interferometric system. In Example 9, polarization is achieved by a polarizer located between the wafer and the spectrometer of the interferometric system. In Example 10, the first and second settings are orthogonal polarizations of the reflected signal. In Example 11, the wafer is circular, and the first and second settings are orthogonal polarizations in the radial and tangential directions of the wafer.

[0079] Example 12 includes features of any of the examples above, and a function that determines a coherent portion of the reflected signal from each of the multiple spectra measured from the first and second groups, including fitting a mathematical equation of the interferometric measurement signal to the reflected signal.

[0080] Example 13 includes features of any of the examples above, and the structure includes one of the following: through-silicon via (TSV), DRAM capacitor, 3D-NAND structure, 3D DRAM structure, CMOS image sensor, or advanced packaging features.

[0081] Example 14 includes features of any of the examples above, and a part of the mathematical equation representing the coherent portion of the reflected signal as a function dependent on the z-interferometric measurement.

[0082] Another example of the present invention includes a metrology unit for measuring optical critical dimensions, the metrology unit comprising: a) An interferometer with adjustable mirror positioning; b) Polarization control system; c) Programmable aperture system; and d) A processor configured to perform the method of any one of claims 1-14.

[0083] The metering unit may also include a mirror support system, which typically enables controlled tilting of the interferometer mirror under the control of the processor.

[0084] Another example of the invention includes a non-transitory computer-readable medium storing instructions for optical critical dimension (OCD) metrology for characterizing a structure by interferometry, the structure having at least two reflective surfaces, including a lower (or bottom) surface separated from at least one higher (or top) reflective surface by a distance greater than the coherence length, the instructions being configured to implement any of the methods of claims 1-14.

[0085] It should be understood that another example of the present invention is a metering unit configured to implement any of the method examples above. Additionally, another example of the present invention is a non-transitory computer-readable medium storing instructions for implementing any of the method examples above.

Claims

1. A method for optical critical dimension (OCD) metrology for characterizing the structure of a wafer using an interferometric measurement system, wherein, The structure has at least one lower surface separate from the higher surface, and the method includes the following steps: a) For each of the plurality of linear positions z of the interferometer mirror of the interferometric measurement system, while applying a first setting of asymmetry-sensitive measurement parameters, the spectra of the interferometer signals reflected from the lower surface and the higher surface are measured to create a set of plurality of spectra, wherein each linear position z is within the coherent optical path difference (OPD) matching range of the lower surface. b) From the set of measured spectra, determine a function representing the coherent portion of the reflected signal, and apply a time-domain transformation to the function to generate a first signal feature representing the coherent portion in the time domain; c) Repeat steps a and b for the second setting of the asymmetric sensitive measurement parameter to generate a second signal feature; d) Calculate a measure of the difference between the first signal feature and the second signal feature, and output the measure of the difference as an indicator of the asymmetry of the structure.

2. The method according to claim 1, wherein, The light emitted from the light source of the interferometric measurement system has a coherence length, and the lower surface is separated from the higher surface by a distance greater than the coherence length.

3. The method according to claim 1, wherein, The measure of the difference is a function of the difference between the first feature and the second feature within the optical path difference (OPD) matching range of the features.

4. The method of claim 1, further comprising determining the severity of the asymmetry by comparing a measure of the difference with a previously calibrated threshold.

5. The method according to claim 1, wherein, The difference between the first signal feature and the second signal feature is measured as the difference between the areas under the curves of the corresponding signal features.

6. The method according to claim 1, wherein, The difference between the first signal feature and the second signal feature is measured as the difference between the signal amplitudes of the corresponding signal features.

7. The method of claim 1, further comprising determining the one or more size asymmetries by comparing a measure of the difference with a previously determined measure of the difference related to one or more size asymmetries of the structure via a function or machine learning model.

8. The method according to claim 1, wherein, The asymmetry-sensitive measurement parameter is one or more of the following: the polarization state of the reflected interferometer signal; the angle of incidence (AOI) achieved by the tilt orientation of the mirror; and the aperture adjustment for partially blocking light emitted at the light source or light received at the spectrometer of the interferometric measurement system. And the offset of the light spot focused on the structure by the interferometric measurement system.

9. The method according to claim 8, wherein, The polarization is achieved by a polarizer located between the wafer and the spectrometer of the interferometric measurement system.

10. The method according to claim 8, wherein, The first setting and the second setting are the orthogonal polarization of the reflected signal.

11. The method according to claim 8, wherein, The wafer is circular, and the first and second settings are orthogonal polarizations in the radial and tangential directions of the wafer.

12. The method according to claim 1, wherein, Determining a function representing the coherent portion of the reflected signal from each of the multiple spectra measured from the first and second groups involves fitting a mathematical equation of the interferometric measurement signal to the reflected signal.

13. The method according to claim 1, wherein, The structure includes one of the following: through-silicon via (TSV), DRAM capacitor, 3D-NAND structure, 3D DRAM structure, CMOS image sensor, or advanced packaging feature.

14. The method according to claim 1, wherein, The function representing the coherent portion of the reflected signal is part of the interferometric mathematical equation that depends on z.

15. A measurement unit for measuring critical optical dimensions, comprising: An interferometer with adjustable mirror positioning; Polarization control system; Programmable aperture system; as well as The processor is configured to perform the following steps: a) For each of the multiple linear positions z of the interferometer mirror of the interferometric measurement system, while applying a first setting of asymmetry-sensitive measurement parameters, the spectra of the interferometer signals reflected from the lower and higher surfaces are measured to create a set of multiple spectra, wherein each linear position z is within the coherent optical path difference (OPD) matching range of the lower surface. b) From the set of measured spectra, determine a function representing the coherent portion of the reflected signal, and apply a time-domain transformation to the function to generate a first signal feature representing the coherent portion in the time domain; c) Repeat steps a and b for the second setting of the asymmetric sensitive measurement parameter to generate a second signal feature; d) Calculate a measure of the difference between the first signal feature and the second signal feature, and output the measure of the difference as an indicator of structural asymmetry.

16. The metering unit of claim 15 further includes a mirror support system that enables controlled tilting.

17. A non-transitory computer-readable medium storing instructions for optical critical dimension (OCD) metrology for characterizing a structure by interferometry, wherein, The structure has at least two reflective surfaces, wherein the bottom surface is separated from at least one top reflective surface by a distance greater than the coherence length, and the instructions are configured to perform the following steps: a) For each of the multiple linear positions z of the interferometer mirror of the interferometric measurement system, while applying a first setting of asymmetry-sensitive measurement parameters, the spectra of the interferometer signals reflected from the lower and higher surfaces are measured to create a set of multiple spectra, wherein each linear position z is within the coherent optical path difference (OPD) matching range of the lower surface. b) From the set of measured spectra, determine a function representing the coherent portion of the reflected signal, and apply a time-domain transformation to the function to generate a first signal feature representing the coherent portion in the time domain; c) Repeat steps a and b for the second setting of the asymmetric sensitive measurement parameter to generate a second signal feature; d) Calculate a measure of the difference between the first signal feature and the second signal feature, and output the measure of the difference as an indicator of the asymmetry of the structure.

Images